Safe Water Storage in Kenya's Modified Clay Pot:
Standardization, Tap Design, and Cost Recovery
by
Suzanne E. Young
B.S. Environmental Engineering Science
Massachusetts Institute of Technology, 2004
Submitted to the Department of Civil and Environmental Engineering
in Partial Fulfillment of the Requirements for the Degree of
Master of Engineering in Civil and Environmental Engineering
at the
Massachusetts Institute of Technology
September 2005
©2005 Suzanne E. Young. All rights reserved.
The author hereby grants to MIT the permission to reproduce and to distribute publicly paper and
electronic copies of this thesis document in whole or in part.
Signature of A uthor: ...........................
. . . . . . .
....... ...............................
epart ent o:.ivi. and Environmental Engineering
June 28, 2005
C ertified B y: ....................
.........................................
Susan Murcott
Lecturer, Department of Civil and Environmental Engineering
Thesis Supervisor
Accepted B y: ......................................................-
:....................
Andrew Whittle
Professor of Civil and Environmental Engineering
Chairman, Department Committee on Graduate Studies
MASSACHUJSETTS INSTITUE
OF TECHNOLOGY
SEP 15 2005
LIBRARIES
Safe Water Storage in Kenya's Modified Clay Pot:
Standardization, Tap Design, and Cost Recovery
by
Suzanne E. Young
Submitted to the Department of Civil and Environmental Engineering
on June 30, 2005 in Partial Fulfillment of the Requirements for the Degree of
Master of Engineering in Civil and Environmental Engineering
ABSTRACT
One of the main components necessary for providing safe drinking water for users who lack
piped water in the home is the ability to safely store it in the home. Users in the Nyanza Province
of Kenya frequently carry water from some distance or purchase vended water and traditionally
store this water in their homes in clay pots. CARE/Kenya, a non-governmental organization
working in conjunction with local women potters and the Centers for Disease Control and
Prevention, modified these clay pots so that they fit the definition of designated safe storage
water containers, which contain the following three characteristics: a narrow mouth, a lid, and a
tap to prevent recontamination. Three pottery production sites were visited in order to document,
analyze, and suggest improvements for the design of the modified clay pots, specifically with
regards to the standardization of the size and shape of the pots, so as to allow simple and
convenient standardized household chlorine dosing, and the tap design and attachment, because
the current tap design is expensive and prone to leaking. The modified clay pots displayed little
variability in pot dimensions, and the 20 liter modified clay pots from the Amilo location showed
a less than or equal to 10% volume variability that is acceptable according to the Centers for
Disease Control and Prevention household chlorine dosing procedure. A reference rope tool
developed to help standardize the size of modified clay pots can be used to train new potters but
is not necessary for experienced potters. The spring-operated plastic tap was found to be the most
promising design to replace the current metal tap design. The material cost of the modified clay
pots at the three sites was determined with some certainty to range from 202-370 KSH
(US$2.70-US$4.90). However, labor costs, transportation costs, profit margins and the role of
subsidies, if they exist, were unable to be accurately determined. Therefore, cost recovery of the
modified clay pot is unclear at all three pottery sites due to insufficient data.
Thesis Supervisor: Susan Murcott
Title: Lecturer in Civil and Environmental Engineering
Acknowledgements
"Do you not know? Have you not heard? The Everlasting God, the Lord, the Creator of the ends
of the earth does not become weary or tired. His understanding is inscrutable. He gives strength
to the weary, and to him who lacks might He increases power. Though youths grow weary and
tired, and vigorous young men stumble badly, yet those who wait for the Lord will gain new
strength; they will mount with wings like eagles, they will run and not get tired, they will walk
and not become weary." (Isaiah 40: 28-3 1)
To God, thank you for your strength, thank you that your power is made perfect in my weakness,
thank you that your grace is sufficient for me.
To Susan Murcott, thank you for finding this project and for your careful reading of this thesis.
To Daniele Lantagne, thank you for your help in kick-starting this project in Kenya, careful
reading of this thesis, quick replies to emails and encouragement throughout.
To CARE Kenya, thank you for hosting me in Homa Bay and making this project possible.
To Eric Adams, Cynthia Stewart, and Blanche Staton, thank you for your administrative help
regarding finishing this thesis during the summer. Dr. Adams, thank you also for the last minute
calculus help!
To AC, EC, and CR, thank you for your help with taking pictures, giving me access to a digital
camera, and giving me access to a scanner.
To my M.Eng classmates, thank you for a fun year.
To my friends, AP, PH, JT, AC, AF, RB, KL, MC, JZ, RW, EH, CBCGB girls group and Charis
home group, thank you for listening to my thesis woes, encouraging me in hard times and
supporting me throughout.
To my family, Mommy, Daddy, SM, KL, and BN, thank you for loving and supporting me 100%,
and for constantly reminding me not to give up and that in the end I will always be okay.
Contents
Chapter 1 General Introduction.....................................................................................
1.1
1.2
1.3
1.4
1.5
Introduction .....................................................................................................
W ater in Kenya..............................................................................................
Safe W ater System ..........................................................................................
M odified clay pot............................................................................................
CARE/Kenya ...... s....
......................................
..........................................
9
9
12
13
14
15
Chapter 2 Research Objectives .................................................................
18
2.1 Research objectives..........................................................................................
18
2.1.1 Standardization ................................................................................
2.1.2 Tap design and attachm ent...............................................................
2.1.3 Cost recovery analysis .....................................................................
Chapter 3 Literature Review ..........................................................................................
3.1 M aster of Engineering theses..........................................................................
3.2 Organization-based reports............................................................................
3.3 Other works.....................................................................................................
Chapter 4 Methodology.................................................................................................
4.1
4.2
4.3
4.4
Observation .....................................................................................................
Interviews............................................................................................................29
Trial-and-error problem solving .........................................................................
Focus groups..................................................................................................
4.4.1 Definition and motivation..............................................................
4.4.2 Form at of discussions .....................................................................
18
19
20
21
21
23
24
28
29
31
31
31
33
Chapter 5 Focus Group Results......................................................................................35
5.1
5.2
5.3
5.4
5.5
Background on shapes of clay pots.......................................................
Preparation for focus groups..........................................................................
Am ilo focus group .........................................................................................
Oriang focus group .........................................................................................
Summ ary
.....................................................................................
.35
38
39
43
46
Chapter 6 Standardization.............................................................................................
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
Background and m otivation............................................................................
Construction of m odified clay pots.................................................................
Pre-existing techniques ...................................................................................
Shrinkage .......................................................................................................
Variability in pot dim ensions and volumes ...................................................
Techniques and tools to standardize pottery ...................................................
Reference ropes...............................................................................................
Summ ary of results ..........................................................................................
Chapter 7 Tap Design and Attachment ........................................................................
7.1
7.2
7.3
7.4
7.5
Current design and attachment ........................................................................
Leakiness from current tap ..................................................................................
Alternative designs..........................................................................................
Plastic tap ............................................................................................................
Summ ary of results .........................................................................................
48
48
50
53
54
61
74
75
79
81
82
88
90
92
99
Chapter 8 Cost Recovery Analysis..................................................................................101
8.1
8.2
8.3
8.4
8.5
Cost breakdown of Am ilo m odified clay pot....................................................102
Cost breakdown of Oriang modified clay pot...................................................107
Cost breakdown of Kinda E Teko (Asembo) modified clay pot ...................... 110
Limitations of analysis......................................................................................112
Summ ary of results ...........................................................................................
113
Chapter 9 Conclusions......................................................................................................114
9.1
9.2
9.3
9.4
Standardization .................................................................................................
Tap design and attachm ent................................................................................115
Cost recovery ....................................................................................................
Overall conclusions...........................................................................................116
References...........................................................................................................................117
114
116
List of Figures
Figure
Figure
Figure
Figure
1-1.
1-2.
1-3.
1-4.
A provincial m ap of Kenya............................................................................
10
W ide-m outh clay pot.....................................................................................
10
The m odified clay pot .....................................................................................
12
Map of Kenya showing administrative boundaries........................................17
Figure 4-1. Map of Kenya's Nyanza Province and field site locations .............................
Figure
Figure
Figure
Figure
28
5-1. Examples of different traditional forms of Kenyan pottery...........................37
5-2. Various shapes of clay pots............................................................................
38
5-3. The author's drawings of pot shapes..............................................................39
5-4. Examples of Amilo Potters' individual drawings of clay pots ...................... 41
Figure 5-5. Examples of Oriang Potters' individual drawings of clay pots .....................
44
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
6-1. Dimensions of the modified clay pot ............................................................
50
6-2. Forming base by pounding clay into a flat disk ............................................
51
6-3. Extending vessel w alls ..................................................................................
52
6-4. Continued vessel wall extension ...................................................................
52
6-5. Before and after sm oothing............................................................................
53
6-6. Rate of water loss in drying moist clay ..........................................................
59
6-7. Stages in drying moist clay ............................................................................
59
6-8. Measuring pots with a seamstress tape at Amilo ..........................................
61
6-9. Graphical representation of the dimensions of 20 L modified clay pots .....
64
6-10. Graphical representation of the dimensions of 40 L modified clay pots ......... 65
6-11. Variability in 20 L modified clay pots made by Amilo Potter No. 7...........67
6-12. Variability in 40 L modified clay pots made by Amilo Potter No. 2...........67
6-13. Symbols used in equations to convert dimensions to volumes .................... 69
6-14. Varying shapes of modified clay pots..........................................................72
6-15. Hardwood calipers and aluminum calipers used to standardize pottery .......... 75
6-16. Pictorial drawing of 20 L modified clay pot reference rope ........................ 77
6-17. Photograph of 20 L modified clay pot reference rope. ...............................
77
6-18. A potter using the 20 L modified clay pot reference rope at Amilo ........... 78
6-19. Teaching the potters to use the reference rope tool .....................................
78
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
7-1.
7-2.
7-3.
7-4.
7-5.
7-6.
7-7.
7-8.
7-9.
Exam ple of a m etal tap...................................................................................
82
Cutting PVC pile to proper length with a hacksaw blade .............................
83
Heating PVC pipe to thermally press fit it to the metal tap threads...............84
The PVC pipe fits snugly over the metal tap threads.....................................84
Close-up of split PVC pipe used to fit tap to PVC pipe..................................84
Tap attachm ent in parallel .............................................................................
85
Tap w edged in vessel .....................................................................................
86
Tap is fixed in place using nill on inside and outside of vessel.....................86
Nill is applied to the base, forming an outer layer .........................................
86
Figure 7-10.
Figure 7-11.
Figure 7-12.
Figure 7-13.
Figure 7-14.
Figure 7-15.
Figure 7-16.
Figure 7-17.
Figure
Figure
Figure
Figure
Figure
8-1.
8-2.
8-3.
8-4.
8-5.
Nill is extended up to the tap and around the base of the vessel..................87
Using rocks and pottery shards to stabilize tap............................................87
Alternative designs to current metal tap design ..........................................
90
Potters for Peace Filtr6n System.................................................................93
Schematic drawing of the spring-operated plastic tap .................................
94
The plastic tap attached to a small flower pot at Oriang..............................95
Close-up of jam nut used to secure the plastic tap .......................................
96
A 20 L modified clay pot at Oriang with plastic tap attached .................... 97
Interviewing Amilo CBO members .................................................................
103
Pie chart showing material cost breakdown at Amilo (Amilo CBO)...............104
Pie chart showing material cost breakdown at Amilo (Mwaki, 2005).............106
Pie chart showing material cost breakdown at Oriang.....................................109
Pie chart showing material cost breakdown at Kinda E Teko ......................... 111
List of Tables
Table 6-1. Shrinkage line tests performed at Oriang ........................................................
55
Table 6-2. Dimensions of modified clay pots freshly made and after drying...................56
Table 6-3. Percent shrinkage of modified clay pots at Oriang .........................................
57
Table 6-4. Dimensions of 20 L modified clay pots ........................................................
62
Table 6-5. Dimensions of 40 L modified clay pots .........................................................
63
Table 6-6. Variability in modified clay pot dimensions by location ................................
66
Table 6-7. Measured volumes and dimensions of four 20 L pots.....................................68
Table 6-8. Comparison of measured volumes to calculated volumes ..............................
70
Table 6-9. Calculated volumes of Amilo 20 L modified clay pots...................................72
Table 6-10. Calculated volumes of modified clay pots by location and size ................... 73
Table 6-11. Dimensions calculated for the 20 L modified clay pot reference rope..........76
Table 7-1. Leakiness experiment at Oriang .....................................................................
Table 7-2. Leakiness experiment at Amilo .......................................................................
Table 7-3. Pugh chart used to evaluate tap design alternatives .......................................
89
89
92
Table 8-1. Material cost breakdown of Amilo modified clay pot (Amilo CBO) ............... 104
Table 8-2. Material cost breakdown of Amilo modified clay pot (Mwaki, 2005)..............106
Table 8-3. Summary of costs and sale prices for Amilo modified clay pots ..................... 107
Table 8-4. Material cost breakdown of Oriang modified clay pot......................................108
Table 8-5. Summary of costs and sale prices for Oriang modified clay pots .................... 109
Table 8-6. Material cost breakdown of Kinda E Teko modified clay pot .......................... 110
Table 8-7. Summary of costs and sale prices for Kinda E Teko modified clay pots ......... 111
Table 8-8. Summary of cost of metal tap and total material costs......................................113
Chapter 1
General Introduction
1.1
Introduction
One of the main components necessary for providing safe drinking water is the ability to
safely store it in homes of users that do not have a piped water supply at all times. Whether the
water is obtained from an improved source, such as a protected well; cleaned, such as by using
filters or a coagulant; or disinfected through SODIS, household chlorination, or some other
method, the potential for contamination in transport and storage undermines and can negate the
resources spent ensuring the water is safe to drink at the improved source.
People in the Nyanza Province of Kenya (Figure 1-1), located along the shores of Lake
Victoria, have a tradition of storing their water in wide-mouth clay pots (Figure 1-2). Users
prefer using clay pots because of the evaporative cooling effect such vessels have on water and
because the clay makes the water palatable (Bovin and Morohashi, 2004). However, the wide
mouths encourage the drawing of water with cups, and often, the hands holding the cups are
contaminated. So, even pristine water stored improperly can be easily contaminated, leading to
incidences of diarrheal disease, mainly in children under the age of five.
9
Figure 1-1. A provincial map of Kenya showing the location of Nyanza Province on the shores of
Lake Victoria (western border).
(Source: United Nations Development Programme-Kenya/ GEF-SGP, 2004)
Figure 1-2. Wide-mouth clay pot.
(Source: CARE Kenya, 2003)
10
One approach to minimizing potential contamination routes in storage is the use of a
designated safe water storage container. Safe Water System (SWS) containers are designed to
eliminate potential routes of contamination when used properly, by providing a fully enclosed
container with a tap that enables safe access to water. The Centers for Disease Control and
Prevention (CDC) defines a safe storage container as having the three following characteristics:
1) narrow mouth; 2) lid; and 3) a tap to prevent recontamination (CDC, 2005).
This study focuses on a safe water storage container called the "modified clay pot"
(Figure 1-3), which was developed by the non-governmental organization (NGO) CARE/Kenya,
working in conjunction with local women potters and the CDC. Despite their improvements to
the wide-mouth clay pot to make it into a safe storage container (narrowing the mouth, adding a
lid, and attaching a tap), the design of the modified clay pot needs further improvement (Quick,
2004). The metal tap leaks, and is very expensive, costing nearly the same amount as all of the
other materials combined. In addition, the size and shape are non-standard. A standard size is
essential for proper chlorine dosing, and a standard shape is ideal to ease tap attachment.
This thesis begins in Chapter 1 with a brief introduction to water in Kenya, the Safe
Water System (SWS), the modified clay pot, and the role of CARE/Kenya regarding the
modified clay pot. The research objectives are stated in Chapter 2, followed by a literature
review in Chapter 3, the methodology in Chapter 4, a spotlight on the focus group results in
Chapter 5, a discussion and results of standardization in Chapter 6, a discussion and results of the
tap design and attachment in Chapter 7, a cost recovery analysis in Chapter 8, and conclusions in
Chapter 9.
11
own
Lid
Narrow mouth
Sediment pouch
Metal tap to access wa
Flat base
Figure 1-3. The modified clay pot.
(Source: Murcott, 2004)
1.2
Water in Kenya
Kenya is a country struggling to meet the water demand of its 31.5 million people. Only
57% of the population has access to an improved water supply (e.g. piped household connection,
public standpipe, borehole, protected dug well, protected spring, or rainwater collection), and
31% of the population has to travel over half an hour to fetch water (Gordon, 2004).
Two thirds of Kenya is arid or semi-arid; therefore, access to water in many areas is very
limited. Women spend several hours daily collecting water, a resource that is shared with
animals that drink from it and contaminate it.
In Nyanza Province, only 14% of people receive their drinking water from a piped supply.
The percentage of people using unimproved open water sources such as springs, streams, and
12
rivers amounts to nearly 58% (Central Bureau of Statistics, 2003). In such open water systems,
the likelihood of contamination is significantly higher than for piped and treated water systems.
One simple and cost effective solution to treating contaminated drinking water and
storing it safely is the Safe Water System (SWS) intervention.
1.3
Safe Water System
According to the Centers for Disease Control and Prevention's (CDC) Safe Water System
web site (2005), "Safe Water System (SWS) is a water quality intervention that employs simple,
inexpensive and robust technologies appropriate for the developing world. The objective is to
make water safe through disinfection and safe storage at the point of use. The intervention
consists of three steps:
1) Point-of-use treatment of contaminated water using sodium hypochlorite solution
purchased by a local manufacturer or in the community;
2) Safe water storage in plastic containers with a narrow mouth, lid, and a tap to
prevent recontamination; and,
3) Behavior change techniques, including social marketing, community mobilization,
motivational interviewing, communication and education. These activities
increase awareness of the link between contaminated water and disease, the
benefits of safe water, and hygiene behaviors, including the purchase and proper
use of the water storage vessel and disinfectant."
Quick et al. (2002) field tested a Safe Water System intervention that consisted of water
treatment, safe storage, and community education in Kitwe, Zambia. They tested a random
selection of 166 intervention households and 94 control households and found that diarrheal
13
disease risk for individuals in intervention households was 48% lower than for controls. Semenza
et al. (1998) also found that the use of SWS reduced diarrheal disease incidence by 85% in
Uzbekistan. For this particular thesis, the focus is safe water storage not in plastic containers as
in Zambia and Uzbekistan, but in clay containers, also with a narrow mouth, lid, and a tap to
prevent recontamination. This container is known as the "modified clay pot".
1.4
Modified clay pot
The modified clay pot (Figure 1-3) includes the following features recommended by the
CDC: a narrow mouth, a lid, and a tap to access water. In addition, the modified clay pot has the
benefits of a flat base for easy water extraction and a space at the bottom (below the tap) to retain
sediment. The form, color, and function of the traditional clay pot are essentially retained.
However, there are several weaknesses associated with the modified clay pot's design. These
include fragility of the structure, leakage around the metal tap, and a non-standard size as a result
of manual molding. Because CARE/Kenya is promoting treatment of water with a 1% sodium
hypochlorite solution at the household level, a standard size is essential to choose the proper dose.
Moreover, from an affordability standpoint, there is interest to replace the metal tap with a
different type of tap because the metal tap itself costs about the same as the entire clay pot.
A brief history of CARE/Kenya's role in conceptualizing and disseminating the modified
clay pot is found in the next section.
14
1.5
CARE/Kenya
CARE/Kenya has been one of the non-governmental organizations (NGOs)
implementing the SWS program in Kenya. CARE is an international NGO that has contributed
to poverty reduction in Kenya for the past 32 years. After conducting a baseline survey in
February 2000 that revealed that 99% of the target population stored their drinking water in
traditional wide-mouth clay pots, CARE/Kenya started a unique adaptation of the SWS
intervention (Mwaki, 2005). Although the second step of the SWS intervention calls for the use
of safe water storage in plastic containers with a narrow mouth, lid, and a tap, CARE/Kenya
found that the target community preferred clay pots because of their cooling effect on water, a
key point considering the hot climate of the region. Common 20-liter plastic jerry cans are used,
however, to collect and transport water from the sources in the target region of Nyanza Province.
In response to the community's preference, CARE/Kenya's SWS team designed a
modified clay pot with features similar to the improved plastic container. This proved a rigorous
exercise, especially the molding of the pot to both incorporate the SWS features and keep a
similar design to the wide-mouth clay pot. During the re-designing of the pot in August 2000, the
SWS team contacted a local pottery group known as the Oriang Women's Pottery Group to make
the modified clay pots on behalf of CARE/Kenya.
When the first batch of pots were ready to be sold at the market, the demand was already
much greater than the limited supply. The features of the improved pots, including the metal taps
and lids, appealed to the local community, increasing demand. At this point, the CARE/Kenya
SWS team recognized the need to scale up the production by identifying additional pottery
groups across the region, thereby decentralizing the production.
15
In 2003, the SWS team identified a local pottery group known as the Amilo Community
Based Organization (CBO) to be trained in the SWS intervention. This group eventually started
manufacturing the modified clay pots in addition to supplying the chlorine for water treatment.
At this same time, the SWS intervention greatly expanded to other districts across the
Lake Victoria region. When CARE/Kenya held training sessions about SWS, one of the
implementing partners, Society for Women and AIDS in Kenya (SWAK), identified a pottery
group named Bar chando Women's Group to make the modified clay pots. The CARE/Kenya
team trained this group and ultimately distributed the pots to the support groups working under
the umbrella of SWAK. The Bar chando Women's Group has made pots that are sold 100 km
away in Kisumu City (Figure 1-4).
Beginning in February 2005, CARE/Kenya's SWS intervention began to expand to fifty
schools in three districts of Nyanza Province: Homa Bay, Suba, and Rachuonyo (Figure 1-4). In
each school, the SWS team earmarked a minimum of five modified clay pots for the storage of
treated drinking water. In addition, the team is making available hand-washing facilities such as
plastic vessels with taps in these schools. In March 2005, the team expanded the SWS
intervention to the Siaya District (Figure 1-4) through the training of all health workers in the
health facilities that provide Prevention of Mother to Child Transmission (PMTCT) of HIV
services. SWS has also been integrated into the existing Water Sanitation and Education for
Health (WASEH) project that is currently implementing activities in two districts as of April
2004. CARE/Kenya has high hopes that the SWS intervention will be further scaled up to
include all of Kenya's Ministry of Health facilities.
16
TE R
KA
SIA
G
VAHIGA
Isum
BONDO
YAI
KE
RACH UONY-
HO
BA
WS.YAMIFGUC
MIGORI
TRANSM/
RUR IA
Figure
1-4. Map of Kenya showing administrative boundaries.
(Source: FAO, 2005)
17
Chapter 2
Research Objectives
2.1
Research objectives
The goals of this thesis are to document, analyze and suggest improvements for the design
and cost of modified clay pots in Kenya's Nyanza Province, specifically with regards to the
standardization of the pots, design of a new tap and cost recovery.
In parallel with the work of this thesis, in the Master of Engineering thesis of teammate
Michael Pihulic, the entire production and manufacture process of modified clay pots, from start
to finish, was documented using a combination of photography, videotape, and interviews
conducted with the potters (Pihulic, 2005a).
2.1.1
Standardization
The first goal of this thesis is to address standardization of the modified clay pot. In order
to implement a disinfection protocol that is simple and easy to use, either a versatile dosing
system based on volume of household water storage container is required (which is complicated
for users to implement) or much preferably, a standardized container should be utilized or
introduced. Given the variation in the size of non-jerry can household water storage containers, it
is not always feasible to design a simple disinfection dosing procedure for different volumes.
The current method used in Kenya is to use a pre-measured dose of a 1% sodium hypochlorite
solution for a known volume of water. The typical jerry can used to collect water in Kenya is a
20 liter volume; thus, a standard size of the pot, either 20 L or 40 L, is critical to ensure that
people do not fill the pots with e.g. 1
jerry cans. A 20 L or 40 L standard size means the
18
chlorine dosing remains effective. A standard pot size also eases transport of the pots to local
markets and to homes. Various methods for standardizing the size of the clay pot during
production have been investigated and field tested in this thesis.
2.1.2
Tap design and attachment
A second goal of this thesis is to suggest an alternative to the modified clay pot's current
metal tap design and attachment. The tap usually costs more than all the other materials required
to make the modified clay pot (e.g. the tap costs 200 KSH out of a total material cost of 370 KSH
at one pottery production site). In addition, however, the technique for attaching the tap also has
not been perfected and often creates leaks that reduce the effectiveness of the containers. The
goal is to design a replacement tap using locally available resources that will significantly reduce
the cost of the safe water storage container and at the same time reduce leaks.
Several alternatives to the current tap design were developed and evaluated using the
following criteria: cost, leakiness, local availability of materials and potential for distribution in
the Lake Victoria region, ease-of-use for young children, ease of construction, social
acceptability, likelihood of contamination, and maintenance. Each design alternative was
evaluated against the current design using a tool called a Pugh chart.
Ensuring that the local populace is comfortable with any new tap design and attachment
technique used is also a high priority.
19
2.1.3
Cost recovery analysis
Cost recovery of all of the materials used to make the modified clay pot is a key
component, because one of the goals of designing a new tap is to address affordability. Cost data
at Amilo was obtained via interviews and later updated with numbers provided via email
correspondence with CARE/Kenya staff Alex Mwaki (2005), while cost data at Kinda E Teko
was obtained by Gibney (2005) via interview. Cost data for Oriang was incomplete at the time of
field work; however, unknown data was estimated using the Amilo data. The goal of the cost
recovery analysis is to create the best product at the cheapest price, which is integral to creating a
financially viable enterprise.
20
Chapter 3
Literature Review
The modified clay pot has been produced for only five years. Except for the work
documenting the production process of the modified clay pot by teammate Michael Pihulic
(2005a), the Best Practicespublication from Nuffic and UNESCO/MOST (2002), A Manual on
the Safe Water System from CARE Kenya (2003) and an unpublished presentation entitled "Safe
Water System as a Home Care Intervention for People Living with HIV/AIDS" given by Alie
Eleveld at the 2004 International Symposium "Household Technologies for Safe Water" in
Nairobi, Kenya, it has never been studied before. Therefore, this literature review examines the
work of previous Master of Engineering (M.Eng) students in the MIT Department of Civil and
Environmental Engineering (CEE) and others on the related topics of ceramic water filter
manufacture and performance and safe water storage.
3.1
Master of Engineering theses
Three students in the CEE M.Eng program at MIT have previously completed theses
relevant to this study. Their work is summarized below, along with brief explanations as to how
it relates to this study.
Chian Siong Low's (2002) thesis documents the production process of the terra cotta disk
filter in Thimi's Potter Square. Low references the Potter's For Peace (PFP, 2001) web site,
which encourages the use of a press mould for consistency in pot size and shape. Like Nepal,
Kenya also has a long and established tradition in ceramic pottery making. It is therefore ideal
21
for the SWS intervention to build on this tradition. Low also touches upon the issue of
standardizing pot size and shape by use of a press mold.
Rob Dies (2003) documents in detail the ceramic water filter production process of
Madhyapur Clay Crafts in Thimi, Nepal, which includes material selection, processing, shaping
and pressing the filter element into a mold, firing, drying and quality control. He found that
materials include clay, water, combustible material (used to increase the porosity of the material
since combustible material will burn off during the firing, leaving behind pores or voids through
which the water will travel, e.g. sawdust, flour), temper, known colloquially as "grog" (nonplastic material used to reduce shrinking/warping and to control porosity to some extent, e.g.
ground bricks), and colloidal silver (used to reduce microbial contamination). Dies also notes
that equipment is a major capital investment for the potter, and sources of energy in developing
countries are usually unreliable or expensive, if at all available. Therefore, Dies found it
preferable to use non-electricity powered equipment. When mixing clay, Dies reports the goal is
to find the optimal combination of clay, water, combustibles and/or grog, which depends on the
properties of materials, the equipment used, and methods of production. He discovered these
optimal clay recipes depend on both experience and trial and error, and that mixing by hand,
which is common in developing countries, is very labor intensive. This detailed documentation
of ceramic water filter production was valuable for teammate Michael Pihulic's M.Eng thesis
(2005a), which focused solely on the production and manufacture of the modified clay pot. In
addition, Kenyan potters experience similar bottlenecks to production, including limited
equipment due to high capital costs and time- and labor-intensive steps such as mixing clay by
hand.
22
Rebecca Hwang (2003) performed extensive field monitoring of the Potter's for Peace
(PFP) ceramic water filter in Nicaragua. Hwang points out several of the manufacturing and
transporting difficulties related to ceramics, which include inconsistent product standardization
and quality control due to variability in the composition of clay in different geographic locations,
and fragility of the ceramics, which causes easy breakage during transport and inhibits efficient
shipping and delivery to distant rural areas. Hwang suggested improving the filter design to
include increasing durability to reduce breakage by changing materials or adding steps to the
manufacturing process, and standardizing the quality of raw materials to guarantee consistency
of quality in the production. The modified clay pot has similar impediments to those of the PFP
ceramic water filter, including lack of standardization and quality control, and breakage during
transport.
3.2
Organization-based reports
Best practices using indigenous knowledge - Africa, a joint publication from Nuffic and
UNESCO/MOST co-edited by Bovin and Morohashi (2002), contains a report entitled, "The use
of locally produced clay pots modified for safe storage of drinking water in the home-a
component of CARE Kenya's Nyanza Health Water Project." This is one of the only reports
exclusively on the Kenyan modified clay pot. Nuffic and UNESCO/MOST indicate that the
purpose of the modified clay pot is to reduce both the recontamination of treated water during
storage and the incidence of water borne disease transmission. These two organizations state that
the modified clay pots serve this purpose in addition to retaining an aspect of traditional culture,
i.e. storing water in clay pots. Moreover, because the practice of making modified clay pots is
based on indigenous knowledge, Nuffic and UNESCO/MOST consider the modified clay pot
23
cost-effective. For these reasons, Nuffic and UNESCO/MOST deem the modified clay pot a
"Best Practice". This report is valuable because it is the first written work documenting the
modified clay pot. It introduces the practice, defines its purpose, details the role of indigenous
knowledge, and describes the achievements, results and concerns to date.
A Manual on the Safe Water System Based on CARE Kenya's Experience in Rural
Western Kenya by CARE Kenya (2003) is a case study based on CARE Kenya's experiences
while conducting a scientific study and implementing SWS in areas where diarrhea prevalence is
high, due to inaccessibility of safe water. CARE completed an entire study on home-based water
chlorination in the Lake Victoria region. SWS intervention is a household-based point-of-use
water treatment system whose component includes three strategies: 1) Water treatment using a
1% sodium hypochlorite solution branded Klorin (renamed Water Guard); 2) Safe water storage
vessels; and 3) Behavior change. CARE Kenya describes the features of the modified clay pot, as
well as its advantages and disadvantages. This report is relevant because modified clay pots are a
specific example of a local adaptation of the SWS intervention.
Safe Water Systems for the Developing World: A Handbook for Implementing
Household-Based Water Treatment and Safe Storage Projects by The Centers for Disease
Control and Prevention (CDC, 2005) is a detailed and comprehensive reference guide for
implementing household-based water treatment and safe storage projects. It serves to better
clarify the definition of SWS and gives many examples of safe storage containers.
3.3
Other works
Daniele Lantagne (2001a, 2001b) reiterates the PFP mission statement, which includes
the "stability and improvement of ceramic production" and the "preservation of cultural
24
inheritance." Lantagne was contacted to provide an assessment of the Potter's for Peace filter to
USAID/Nicaragua, including information on the intrinsic effectiveness in the laboratory and
field performance in rural communities. Lantagne produced two reports, one on the intrinsic
effectiveness of the filter (including studies on colloidal silver, flow rates, sliver concentration in
water, and the health effects of ingesting silver), and the other on the performance of the PFP
filter under field conditions in six rural communities. Lantagne's survey of 24 families indicated
that the most common problems were the breakage of the filter and recontamination of the stored
water after filtration, and that there currently exists a lack of education about safe water sources
and correct filter cleaning and maintenance in the families using the filter. Lantagne's
recommendations include dedicating a PFP staff member or a local community leader to the role
of following up on filter use in the families, educating families about proper filter use and
maintenance, and developing and selling a cleaning kit with the filters. This study is useful
because it gives an example of the methodology, results, and recommendations of field-testing a
ceramic product. The recommendations may also be relevant to the modified clay pot, i.e.
following up on modified clay pot use in households, educating families about proper chlorine
dosing, and developing and selling cleaning kits with the pots.
Stuart Cheeseman (2003) documented Reid Harvey's methods for producing ceramic
candle filters and ceramic disk filters. Cheeseman catalogs the requisite tools and techniques for
producing these filters. He includes schematics of the presses and molds used to create the filters,
as well as the methods used to produce them. Cheeseman examines the variables inherent in each
stage of the production process and how the local manufacturer, Hari Govinda, has altered and
adapted the production process to produce ceramic filters according to the Reid Harvey method.
These variables include production difficulties found at each stage and how they influence the
25
final product. Cheeseman identified six design issues that were not addressed in the design
specifications provided by Harvey. Cheeseman noted a smearing effect during extrusion that
reduces filter porosity, the necessity of oil or paper to aid in mold extraction, the importance of
obtaining a reliable silver source, the importance of drying following colloidal silver application,
the parameters necessary for achieving good seals, and problems associated with carbon trapping
when firing the ceramics. This study, in common with the theses of Low (2002) and Dies (2003),
provides valuable information on ceramic water filter production methods and tools.
Alie Eleveld, a representative of the CDC/Kenya's Global AIDS Program, presented
"Safe Water Systems as a Home Care Intervention for People Living with HIV/AIDS" at the
2004 International Symposium "Household Technologies for Safe Water," hosted by the Kenya
Ministry of Health, Nursing Council of Kenya, Emory University, Population Services
International (PSI), the Centers for Disease Control and Prevention (CDC) and CARE/Kenya in
Nairobi, Kenya on June 16-17, 2004. Eleveld, working with the Society of Women and AIDS in
Kenya (SWAK) helped to identify local potters in two areas in Asembo to be trained on
manufacturing modified clay pots for safe water storage. Eleveld reported that the first modified
clay pots from the Asembo pottery sites would be available in June 2003.
Teresa Yamana (2004) performed a paper survey of safe storage vessel types in the
Dominican Republic. She found the most likely reason for contamination of stored water is
contact with contaminated human hands or utensils used to scoop water from the storage vessel.
As mentioned previously, contaminated hands are also the main reason for contamination of
water stored in wide-mouth clay pots.
26
Significant research has been completed by MIT CEE M.Eng students, organizations, and
others that is relevant to this thesis. A number of these authors identify some of the same issues
with ceramics as they relate to "safe water" and as seen in Kenya, such as bottlenecks to
production, the role of indigenous knowledge, standardization, breakage, and lack of quality
control. The background literature also places the modified clay pot into the larger context of the
household drinking water treatment and safe storage' intervention. This review of literature
informed the development of the thesis plan and was valuable in fitting this one thesis into a
larger picture.
1http://www.who.int/householdwater
27
Chapter 4
Methodology
Each of the following organizations (with extent of visit in parentheses) was visited by
the author Suzanne Young and teammate Michael Pihulic, with assistance also provided by
Centers for Disease Control and Prevention (CDC) staff and former MIT Master of Engineering
student ('01) Daniele Lantagne: Oriang Women's Pottery Group in Oriang (5 days); Amilo
Community Based Organization (CBO) Pottery Group in Rangwe (5 days); and Kinda E Teko
Pottery Group in Asembo (1 day). These locations are shown on Figure 4-1. While both the
Oriang Women's Pottery Group and the Amilo CBO Pottery Group are supported by
CARE/Kenya, the Kinda E Teko group is supported by the Society of Women and AIDS in
Kenya (SWAK).
Figure 4-1. Map of Kenya's Nyanza Province and field site locations.
(Source: Kenya Web, 2001; United Nations Development Programme-Kenya/ GEF-SGP, 2004)
28
Methodology in the field during January 2005 included observing potters at work,
interviewing potters and project partners, trial-and-error problem-solving with the women
potters, and conducting focus groups. These methods are described in detail in sections 4.1-4.4.
An essential part of the field methodology was working with the women potters to try new
techniques and designs and to think outside of the box.
4.1
Observation
The first day at each site was spent jointly with teammate Michael Pihulic observing
potters at work and visually documenting the current production and manufacturing processes.
Since little was known about the exact processes, it was important to fully understand how the
pots are made, step by step, and to begin to identify potential areas for improvement. This time
was also spent becoming familiar with the potters' skill levels and willingness to try out new
ideas. Observation and visual documentation continued during the remainder of the days spent at
each site. (See Pihulic, 2005a for this complete process documentation.)
4.2
Interviews
The author carried a field notebook and pen at all times in order to record answers to any
questions that came to mind as she observed the potters working. Because the potters only speak
the local dialect of Luo, it was necessary to interview and ask them questions via a hired
translator. (See sections in the next chapter, 5.3 and 5.4, for more detailed information regarding
the translators.) Information on the subject of cost recovery was obtained by interviewing not
only the potters, but CARE/Kenya staff and members of the Amilo Community Based
Organization (CBO) as well. Answers to questions were carefully recorded in field notebooks
29
and reviewed daily to ensure complete understanding of the responses. Any vagueness was
clarified the next day with follow-up questions.
Here is a sampling of some of the questions asked and answers gathered during
interviews of the potters (via a translator) and other members at the Amilo CBO:
Q: How long have the potters been making pottery,
and specifically, modified clay pots?
A: The potters have each been making pottery for about ten years; they have made
modified clay pots since 2003.
Q: How did the potters
learn how to make pottery, and specifically, modified clay pots?
A: Each potter learned how to make pottery from her mother or mother-in-law. Two
Amilo potters were sent to Oriang to learn how to make the modified clay pots and
subsequently passed on this knowledge to other Amilo potters.
Q: What is the production rate of modified clay pots?
A: The average potter can make up to three pots per day; however, production rates are
limited by potters having to share equipment and tools.
Q: What techniques have the potters tried to standardize
the modified clay pot?
A: Potters rely on the length of their arms and the width of their hands as well as
experience to shape modified clay pots into 20 L and 40 L sizes.
Q: For what price
are the modified clay pots sold at the market?
A: 20 L modified clay pots are sold for 380 KSH; 40 L modified clay pots are sold for
500 KSH.
Q: Do the potters test the
pots for leakage before selling them?
A: No, no attempt is made to test the pots before selling them.
30
4.3
Trial-and-error problem solving
After observing and interviewing potters at the field sites, and identifying potential areas
of improvement to the modified clay pot production process, the author engaged daily in
brainstorming sessions with teammate Michael Pihulic to formulate new ideas and decide which
of these ideas to test in the field with the potters. Brainstorming sessions were conducted by
discussing ideas aloud, posing possible solutions to problems identified, and evaluating each idea
in terms of available time, access to materials, feasibility, and potters' willingness to participate
in trying any new ideas developed. Reference books on ceramics properties and fabrication
processes were also consulted to flesh out ideas in Kenya. If necessary, calculations were worked
out and tools were constructed in preparation for the next day in the field. This approach of trialand-error problem solving proved very effective and was used to address both the issue of
standardizing the pots and designing a new tap.
4.4
Focus groups
In order to garner the local women potters' thoughts about the origins of the shape of the
modified clay pot, and to have them develop ideas for pot shapes that would better ease the
insertion of a tap, i.e. a pot that has at least one flat area large enough to accommodate a tap will
create a fit between tap and pot that is less prone to leaking, focus groups were conducted at both
Amilo and Oriang.
4.4.1
Definition and motivation
Focus groups consist of about 6 to 10 people that are part of a relevant target market or
political entity gathered in the same room and guided through a moderator-led discussion that
31
investigates attitudes about a specific client's services, products, or political persuasions. While
often used in market research or political analysis, focus groups can also simply be a small group
representative of a wider population and guided though open discussion for its members'
opinions about or emotional response to a specific area or subject.
Focus groups typically have six features, according to Focus Groups: A Practical Guide
for Applied Research by Richard A. Krueger (1994): "(1) people, (2) assembled in a series of
groups, (3) possess certain characteristics, and (4) provide data (5) of a qualitative nature (6) in a
focused discussion." Krueger (1994) further elaborates on each of these six features:
(1)
Focus groups are typically composed of 6 to 10 people, but the size can
range from as few as 4 to as many as 12. The size is conditioned by two
factors: It must be small enough for everyone to have opportunity to share
insights and yet large enough to provide diversity of perceptions.
(2)
The focus group interview is conducted in a series. Multiple groups with
similar participants are needed to detect patterns and trends across groups.
(3)
Focus groups are composed of people who are similar to each other [and]
who do not know each other.. .More recently, however, researchers are
questioning the necessity and practicality of this [latter] guideline,
especially in community-based studies.
(4)
Focus groups produce data of interest to researchers. In this respect the
purpose differs from other group interactions in which the goal is to reach
consensus, provide recommendations, or make decisions among
alternatives.
(5)
Focus groups produce qualitative data that provide insights into the
attitudes, perceptions, and opinions of participants. These results are
solicited through open-ended questions and a procedure in which
respondents are able to choose the manner in which they respond and also
from observations of those respondents in a group discussion. The
researcher serves several functions in the focus group: listening, observing,
and eventually analyzing, using an inductive process.
(6)
The topics of discussion in a focus group are carefully predetermined and
sequenced.
32
Focus groups were included in the methodology because it was ideal to ask questions and
encourage discussion in a group setting. Focus groups provided a dynamic atmosphere for
participants to share ideas. Because the potters have the first-hand experience in producing and
manufacturing the modified clay pots, it is their attitudes, beliefs and perceptions that are of
interest. Discussions in a group setting were also more time efficient than one-on-one interviews.
4.4.2
Format of discussions
The author's objectives of the focus group discussions at each site included the following:
-
Discover the potters' thoughts on the origins of the rounded shape typical of the
wide-mouthed and modified clay pots;
-
Obtain the potters' opinions about a cylindrical shaped pot (an example of which
was made the previous day according to the author's specifications at the Amilo
CBO);
-
Have the potters each come up with her own drawings of pot shapes after being
told that the goal is to ease insertion of the tap (i.e. because the tap is currently
inserted on the rounded surfaces of the pots, the "fit" is not very good-a flat
surface is ideal); and,
-
Determine preferences of several different sketches of pot shapes drawn by the
author, also keeping with the condition that the aim is to ease insertion of the tap.
33
The only deviation from the objectives sought during the Amilo focus group versus those sought
during the Oriang focus group is that the researchers did not ask the Oriang potters to make a
cylindrical shaped pot at all, so the question of whether they preferred the rounded shaped pot or
the cylindrical pot (i.e. their opinions about a cylindrical shaped pot) was not asked.
The results of observation, interviews and trial-and-error problem solving are found
throughout Chapters 6, 7, and 8, while the results of the focus group discussions are found in
Chapter 5.
34
Chapter 5
Focus Group Results
Before the results of the focus groups conducted at Amilo and Oriang are discussed, it is
important to understand the variations in the shapes of clay pots in traditional Kenyan pottery
and how the skills and knowledge of pottery making are passed down.
5.1
Background on shapes of clay pots
In western Kenya, pottery has almost exclusively been in the woman's domain
(Mutagaywa, 1995). However, only a small proportion of the women in any potting community
are involved in the trade, and those women who do make pottery usually undertake this craft
part-time in combination or alternating with other household tasks (Wandibba 2003).
Women produce various types of clay pots and pans mainly for household use in food
processing, preparation, and storage and for water carriage and storage. Mutagaywa (1995)
classifies traditional Kenyan pottery into three forms (with Luo terminology in parentheses):
1) Simple open form: Smaller sizes for processing and serving meat and vegetables, or
larger sizes for brewing and serving beer or fermenting and serving porridge of
millet/maize (nyalora);
2) Simple restricted form: For cooking and preparing fish (haiga),or for cooking and
storing grains and starchy foods (kabange or dakuon);
3) Forms with a neck: For brewing and storing beer (mbiru), for carrying and storing
water (dapi), or for odd jobs (agulu).
35
Figure 5-1 shows examples of some of these different traditional forms of pottery. Those
in the foreground (haiga)are for cooking fish, while those with a neck (dapi) are for carrying
and storing water. Examination of different pottery forms produced suggests that variation in size
within form classes relates to differential function, such that potters distinguish these differences
with terminological distinctions (Wandibba 2003). However, sometimes distinct forms have the
same terminology. One specific example of this is described by Wandibba (2003):
The Okiek [tribe of west central Kenya] traditionally made three different forms
of the honey pot, depending on whether the honey pot was to be stored in the
forest or at home. Although the three forms were ingrained in the potter's mental
template, she had no separate names for them and simply referred to all three as
"honey pots." Although all three forms had necks, the pots varied in size and body
form. The vessel for storing the honey in the forest was elongated in form and had
a narrow mouth to accommodate a stopper to facilitate sealing before the pot was
buried in the ground for safekeeping. The shape designed for storage in a cave
was much bigger and had a rounded body to enable it to stand unsupported on a
shelf or the floor. Finally, the house honey pot (the only one of the honey pots still
being made today) has a rounded body and a short cylindrical neck, with a tightfitting leather cover and leather carrying handle. It would thus appear that unique
functions produce subtle differences in vessel forms.
This example explains that one function of the rounded body is to enable a pot to stand
unsupported on a shelf or floor, and that rounded bodies can be much bigger than
elongated forms. This suggests why pots used for water transport and storage
traditionally have rounded bodies, i.e. rounded pots are bigger and can hold more water.
36
Figure 5-1. Examples of different traditional forms of Kenyan pottery with different uses; those in
the foreground (haiga)are for cooking fish, while those with a neck (dapi)are for carrying and
storing water.
(Source: Mutagaywa, 1995)
The wide-mouth clay pot (Figure 5-2a), the traditional shape of pots used to store
water in the Nyanza Province, has a rounded body, a wide mouth, and a rounded base.
The method for drawing water from such shaped pots typically involves dipping in cups
or calabashes with one's hands. The modified clay pot (Figure 5-2b), a variation of the
traditional shape, has a more prominent base so that sediment can settle below the tap, a
rounded body, a narrow neck opening to a narrow mouth, and a lid, also made of clay.
This shape, which is a designated safe water storage container, was developed by CAREKenya's SWS team in 2000. Water is dispensed by turning the tap handle. An alternate
pot shape, the cylindrical pot (Figure 5-2c), was suggested by the author and built by one
of the potters at the Amilo Community Based Organization (CBO). The idea behind this
shape is that a flat, not round, surface would ease insertion of the tap and reduce leakiness
around the tap caused by an imperfect fit between the clay pot and tap.
37
a
b
c
Figure 5-2. Various shapes of clay pots: wide-mouth clay pot (a), modified clay pot (b), and
cylindrical pot (c).
(Source: CARE Kenya, 2003; Murcott, 2004; Pihulic, 2005b)
Skills and knowledge of pottery making are passed through generations (mother to
daughter, mother-in-law to daughter-in-law, aunt to niece, and so on) as well as between the
different regions and tribes via intermarriage, through informal channels of communication such
as observation, imitation and working together (Wandibba, 2003; Mutagaywa,1995).
5.2
Preparation for focus groups
In preparation for the focus group discussions, during a brainstorming session, the author
drew several additional pot shapes (Figure 5-3), keeping in mind that having at least one flat
surface on the pot would aid in tap attachment and reduce leakiness. One of the drawings is a
modified clay pot with an even more prominent cylindrical base (Figure 5-3a), such that the tap
could be attached to this cylindrical portion while still leaving room below the tap for sediment
to settle out. The other drawings included a cylindrical shape with a wider base (Figure 5-3b),
making the height shorter than the cylindrical pot shown in Figure 5-2c; a cylindrical pot with a
38
base that flares out like a salt-shaker (Figure 5-3c); an asymmetrical pot with one rounded side
and one flat side (Figure 5-3d); and a hexagonal pot (Figure 5-3e). Once the author was prepared
for the focus groups with the necessary background research and materials, focus groups were
conducted at Amilo and Oriang.
a
b
c
d
e
Figure 5-3. The author's drawings of pot shapes: rounded pot with extended cylindrical base (a),
cylindrical pot (b), cylindrical pot with flared base (c), asymmetrical pot with one rounded side and
one flat side (d), and hexagonal pot (e).
5.3
Amilo focus group
The first of two focus group meetings was conducted on the third day of field research,
January 8, 2005, at the Amilo CBO. Members of the group included four local women potters
(Amilo Potter #1-4), a translator, a listener from the Centers for Disease Control (CDC), and the
author, who served as the moderator for the discussion. All of the women are of Luo ethnicity
and are mothers. Most of the women learned how to make clay pots from their mothers or their
husband's mothers after they were married, and have roughly ten years experience each (the
range of experience is unknown, as each woman stated they had ten years experience making
pots). However, only two of the women potters traveled to Oriang to receive training in making
39
the modified clay pots. They subsequently taught the other potters at the Amilo CBO the new
design and techniques they learned. It is noteworthy that the translator is an elderly man who is
the General Chairman of the Amilo Community Learning Resources Centre (CLRC), one
component of the Amilo CBO. This position is the most senior position of the Amilo CBO's
Executive Committee. It was originally suggested by the CARE/Kenya representatives upon the
initial meeting between the Amilo group and the researchers that perhaps a young local woman
would be most appropriate to serve as translator. The listener from the CDC, Ms. Daniele
Lantagne, served as an advisor and friend to the research project.
The potters could not say much about the origins of the rounded shape typical of both the
wide-mouth and modified clay pots. (Studies have suggested that shapes are dictated by function,
and that rounded pots used for water transport and storage are shaped so they can hold more
water. See Section 5.1.) The shape is just the way that they were taught to make clay pots by
their mothers or other relatives when they first started. As for whether they preferred the rounded
shape or a cylindrical shape, the women collectively agreed (as summarized by the translator)
that they prefer the cylindrical shape because it is more easily carried, it uses less clay, and
attaching the tap is easier. While the potters themselves support the cylindrical shape, since the
idea is new and people are free to make their own choices, they do not know if the local
population will as readily welcome the differently shaped pot. Therefore, they said they will
make both the modified clay pot and the cylindrical pot and see which one sells better in the
market.
Examples of the potters' individual drawings of clay pots are shown in Figure 5-4. All of
the women's drawings included at least one clearly cylindrical shaped pot (Figures 5-4a, g, h, 1,
40
n), and one or more rounded shaped pots. The women at first were hesitant to start drawing and
needed encouragement from the author and translator. However, it was later evident by their
facial expressions and comments after completing the exercise that they found it enjoyable.
Amilo
Potter #1
a
b
c
d
Amilo
Potter #2
e
f
g
Amilo
Potter #3
h
Amilo
Potter #4
k
I
M
n
Figure 5-4. Examples of Amilo Potters' individual drawings of clay pots.
41
When next asked which of the author's drawings (Figure 5-3) they liked the best, keeping
in mind that they should choose the drawing that would best ease insertion of the tap, two of the
potters, Amilo Potter #1 and Amilo Potter #4, chose the rounded pot with the extended
cylindrical base (Figure 5-3a) because it is easy to attach a tap and particles in the water can
settle out and be retained in the base, below the tap. Amilo Potter #2 originally picked the
cylindrical pot with the flared base (Figure 5-3c) but seemed to be convinced by the group and
translator otherwise. She eventually also chose the rounded pot with the extended base (Figure 53a). Amilo Potter #3 chose the cylindrical pot (Figure 5-3b) because she said making that shape
is easier and that it uses less clay.
After voting on which of the researcher's drawings they preferred most, Amilo Potter #1
proceeded to make the rounded pot with the extended cylindrical base and Amilo Potter #3 a
cylindrical pot with a wider base than was made the previous day. Amilo Potter #1 did not make
the extended base envisioned by the researcher; instead, she just made it less slanted and more
gradual than the base of the modified clay pot (similar to the shape of a clay flower pot). Both a
lack of sufficient communication and translation and the potter's misunderstanding contributed
to this outcome.
There were several limitations of the January 8, 2005 Amilo focus group. One important
thing to note is that a cylindrical shaped pot was made the day before the focus group took place,
as per request by the author. The timing may have been poor because it may have influenced the
potters' preference for the cylindrical shaped pot over the rounded shaped pot. The potters may
also have chosen it because they thought it was the author's own preference. In addition, one of
the reasons they preferred the cylindrical pot is because it uses less clay; however, the amount of
clay was never formally quantified, so whether or not this assumption is true is unknown. Also,
42
whether the women carry through with actually making both the modified clay pot and the
cylindrical pot and selling them at the market remains to be seen. Furthermore, it is believed that
the male translator added bias to the all women's group's responses because he seemed to
dominate their discussion and not merely translate but add his own ideas and convince the
women of certain responses. To what extent this bias contributed is unknown.
5.4
Oriang focus group
A second focus group was conducted at the Oriang Women's Pottery Group on the fourth
day of field research at that site, January 12, 2005. Members of the group included seven local
women potters (Oriang Potter #1-7), a translator, the author (who served as the moderator for the
discussion) and teammate Michael Pihulic. All of the women potters in this group are of Luo
ethnicity and are mothers, each with more than ten years of experience doing pottery work. The
translator was a 25-year old woman who works at a local elementary school teaching
kindergarten-aged children. She was asked to serve as the translator because of her fluency in
both Luo and English, and because she lives nearby and is acquainted with one of the woman
potters.
When asked why pots that store water are molded into the traditional rounded shapes, the
potters said the rounded pots are so shaped in order to better fetch water, i.e. pots with rounded
bodies can hold more water. Historically, rounded ceramic pots were used to both fetch and store
water. They also said that different pot shapes serve different purposes, which is consistent with
background research. Each potter drew shapes of pots, taking into account that a flat spot is
needed to better accommodate tap attachment. Examples of the potters' individual drawings are
shown in Figure 5-5.
43
Oriang
Potter #1
15
a
b
C
d
e
Oriang
Potter #2
-A,
Oriang
Potter #3
't
L'
,
~
h
i
g
Oriang
Potter #4
V
~-
L
I
-
Oriang
Potter #5
m
Oriang
Potter #6
Oriang
Potter #7
p
n&
O
q
a
r
U
Figure 5-5. Examples of Oriang Potters' individual drawings of clay pots.
44
All of the drawings were of rounded pots, except for two single sketches of pots with
flatter surfaces, one drawn by Oriang Potter #1 (Figure 5-5b) and the other by Oriang Pottery #6
(Figure 5-5p). The lack of cylindrical shaped pots is perhaps expected. Recall that at Amilo, the
idea of a cylindrical pot was introduced by the author prior to the focus group discussion;
however, this was not the case at Oriang. It is noteworthy that Oriang Potter #7 labeled her pot
drawings (Figures 5-5r, s, t, u) with their Luo names. Figure 5-5r is a drawing of an ohigla,
which by its shape may be another name for a haiga, a pot used for cooking and preparing fish.
Figure 5-5s is a drawing of a dapi, historically used for water carriage and storage, while Figure
5-5u is a drawing of a dapi tap, which is the name given for a water storage container with a tap,
i.e. the modified clay pot. Figure 5-5t is a drawing of a kabange, used for cooking and storing
grains and starchy foods.
Finally, potters were shown the researcher's drawings of various pot shapes (Figure 5-3)
and asked to choose which they liked best by raising their hands. Six potters chose the rounded
pot with the extended cylindrical base (Figure 5-3a), while two voted for the asymmetrical pot
with one rounded side and one flat side (Figure 5-3d). (Note that an eighth potter joined the
group late and participated in this activity). The potters who chose the rounded pot with the
extended cylindrical base (Figure 5-3a) said they liked it because they have tried that idea before,
and that they liked the rounded, symmetrical shape because it is aesthetically pleasing. The two
potters that chose the asymmetrical pot (Figure 5-3d) said they liked it because they think they
are capable of making it. Furthermore, the potters all said that the shape of the pot does not
matter so much as the cooling effect the clay has on the water and the taste of the water from the
pot.
45
One limitation of this focus group discussion is that the moderator did not have the
complete attention of the members because they were engaged in other activities (e.g. molding
pots, mixing clay) while answering questions. However, unlike at Amilo, the translator at Oriang
was a woman and a teacher (and not an "official"). It is believed that the translation at this site
was more accurate and more effective than at Amilo, and that the translator added little if any
bias to the potters' responses.
5.5
Summary of findings
While not only serving to achieve the objectives sought by the author, the focus groups at
Amilo and Oriang also aimed to encourage the woman potters to try new things and think outside
of the box by having them each participate in drawing pots, hopefully encouraging them to take
ownership of their ideas. A summary of the findings of the focus groups is found below.
-
While the Amilo potters do not know the origin of the rounded shape typical of
both the wide-mouth and modified clay pot, the Oriang potters say that rounded
shapes have historically been used to fetch and store water because they can hold
more water than other shapes. The author's own research into the history of
traditional Kenyan clay pots revealed that rounded bodies enable a pot to stand
unsupported on a shelf or floor and can be much bigger (i.e. can hold more water)
than elongated forms (Wandibba, 2003).
46
-
The Amilo potters collectively preferred the cylindrical pot because they say it is
more easily carried, uses less clay than the rounded pots, and eases tap attachment.
The Oriang potters were not asked to compare the rounded and cylindrical pots.
-
The Amilo potters drew a mix of both cylindrical and rounded pots, while the
Oriang potters drew mostly rounded pots.
-
At Amilo, three potters voted for the modified clay pot with a prominent
cylindrical base, and one voted for the cylindrical pot. At Oriang, six potters voted
for the modified clay pot with a prominent cylindrical base, and two voted for the
asymmetrical pot with one rounded side and one flat side. Overall, the potters
seemed to want to retain the rounded shape of the pots.
Furthermore, the translator plays a key role in the accuracy and effectiveness of the focus
group discussions. While the "ideal" focus group encourages the free flow of ideas, the translator
may inhibit this flow by imposing his or her own views onto the members of the focus group,
thereby preventing the emergence of unbiased responses. It is therefore critical to consider what
type of translator (gender, age, position, etc.) is most appropriate for the focus group members.
Also, the variety of drawings made by women at Amilo differed from those at Oriang, largely
because the idea of a cylindrical pot was introduced at Amilo a day prior to the focus group,
while it was not introduced at Oriang. Lastly, because only two focus groups were conducted, the
discussions and conclusions for this research paper cannot necessarily and should not be applied
to the country of Kenya, or even the Lake Victoria region, as a whole. Patterns and trends across
groups should only be detected after conducting multiple focus groups with similar participants
(Krueger, 1994).
47
Chapter 6
Standardization
In this chapter, the background and motivation for addressing the issue of standardizing
the size and shape of modified clay pots are given. It is essential to understand how pots are
constructed before describing any pre-existing techniques used by potters to address
standardization. In addition, any techniques used for standardization should account for
shrinkage that occurs during drying. The author documents the variability in pot dimensions and
volumes of completed 20 L and 40 L modified clay pots at all three pottery sites and introduces a
reference rope tool to aid potters in standardizing certain dimensions of the modified clay pot.
6.1
Background and motivation
The size and shape of the modified clay pot are currently non-standard, as each potter
makes her own pots following only the rough design. Standardization of the size of the pots is
essential for proper chlorine dosing, while standardization of the shape of the pots is ideal for
prevention of leaking from the tap attachment and prevention of breaking during transportation.
Potters currently produce two different sizes of modified clay pots: roughly 20 liters and
roughly 40 liters. Plastic 20 liter jerry cans are used to collect and transport water from the
source to the household storage container. Thus, from a chlorine dosing perspective, it is
important that modified clay pots are either 20 liters or 40 liters so that people do not fill the pots
with, for example, 1 1/2 jerry cans of water, since chlorine dosing is prescribed for 20 liter volume
multiples. From a practicality standpoint, it is ideal to restrict pot size to hold a maximum of 40
liters. Any size larger than this will increase difficulty in transportation. Additionally, because
the pots will need to be cleaned periodically by the user, the inner mouth diameter and height of
48
the pot need to be appropriate to enable proper washing, even by school children. However, the
mouth size should not be so big as to encourage hand and cup dipping.
The rounded shape of the pot is such that the modified clay pot is wide on the sides, and
short in height. If the pot is too short, the structure will be too weak and the pot will collapse
downward and cave in, or "buckle down". Some members of the CARE/Kenya staff say that
because pots are hand-molded, the shape largely depends on a potter's ability to control the
dimensions of the modified clay pot (Figure 6-1), which include the following:
-
Inner Mouth Diameter
-
Mouth Circumference
-
Middle Circumference (i.e. the pot's widest circumference)
-
Base Circumference
-
Mouth Inflection Point
-
Height
-
Height to Middle Circumference
The height of the Mouth Inflection Point marks the volume of water, either 20 liters or 40 liters,
which a pot can hold. Sometimes, due to poor technique, pots are asymmetrical. The
CARE/Kenya staff also like to joke that potters may have a tendency to make pots like their own
body shapes, e.g. shorter and wider, or taller and narrower; similarly, those purchasing pots may
have a psychological attachment to certain pots, i.e. they prefer to buy pots that are similar to
their own body shapes.
49
Mouth
Mouth Infle
Height
Height to
Middle
Circumference
Circumference
Figure 6-1. Dimensions of the modified clay pot.
Before making any recommendations to help standardize both the size and shape of the
modified clay pots, it was first necessary to learn how pots are constructed and to determine
techniques already utilized by the potters to address standardization.
6.2
Construction of modified clay pots
Modified clay pots are constructed using a handbuilding technique known as segmental
coiling, which is a shaping process suited for producing large storage vessels. Handbuilding
means constructing pottery from pre-made parts that are molded, coiled, or fashioned by hand.
Segmental coiling consists of building each layer of the outer wall of the vessel from several
small ropes, or coils. Individual coils are formed by hand squeezing and rolling the clay into long
ropes. The diameters of the coils should be 2-3 times the desired thickness of the vessel to
50
provide sufficient material to extend the vessel upward while merging with the previous layer.
The coils are then joined to the vessel by pinching, which allows a larger binding area than
simply stacking them and creates a stronger bond. The junctures and seams that are created
between successive coils are then removed by smoothing the surface (Rice, 1987).
Pihulic (2005a) details the process of shaping modified clay pots using the segmental
coiling technique:
The potter begins by taking a fistful of clay and pounding it flat with the palm of
her hand (Figure 6-2). Once the clay has spread out to a certain size the potter
then pinches the edges, forming the beginning of the vessel wall. Once the base
has been formed, the potters begin to form coils of clay by squeezing the clay into
a cylinder shape, and then rolling it to smooth the edges (Figure 6-3). The coil is
then applied to the exposed edge of the vessel, and then pressed on, forming a join.
The material is then pinched and extended upward, and the process is repeated
(Figure 6-4). Once the vessel is roughly half finished in height, the potter
smoothes the interior and exterior by wetting the surfaces and scraping with [a
smoothing tool called] an ombasa. Figure 6-5 [shows] the surface before and after
smoothing. Once the surface has been treated, the potter continues building up the
vessel wall, varying the coil size to control the circumference. The size of the
opening of the neck [is] dictated by the necessary size to give the potter access to
the inside. Because there are no set references or specifications for the dimensions
of the pot they tend to be left to the discretion of the potter.
Figure 6-2. Potter, front, begins forming base by pounding clay into a flat disk.
(Source: Pihulic, 2005a)
51
Figure 6-3. Potter, front, begins extending vessel walls.
Source: (Pihulic, 2005a)
Figure 6-4. Continued vessel wall extension. Coiling process is repeated.
(Source: Pihulic, 2005a)
52
Figure 6-5. Before and after smoothing. Top photo shows discrete layering from coils, while bottom photo
shows the obliteration of distinct layers.
(Source: Pihulic, 2005a)
6.3
Pre-existing techniques
Most of the techniques used to make clay pots, whether wide-mouth or modified, are
passed-down knowledge inherited from foremothers. The potters learn to use their arm length to
judge how tall to make a 20 liter or 40 liter pot, while they use their arm circumference or hand
width to judge how wide to make the mouth diameter. These "standard arm lengths" differ from
potter to potter and are thus relative measurements. The only example of a potter using an
53
absolute measurement scale was found at the Oriang Women's Pottery Group. One of the potters
there uses a measuring tape while molding her pots. She hung it around her neck and used it a
couple of times to measure the pot height and mouth diameter. She claims that in order to make a
20 liter pot, she makes the height 19 inches (48.26 cm), which shrinks to 18 inches (45.72 cm)
after drying, and the mouth diameter 6.5 inches (16.51 cm), which shrinks to 6 inches (15.24 cm)
after drying. Shrinkage is discussed further in the next section.
6.4
Shrinkage
The drying process is an important step in the manufacture of clay articles. Because
shrinkage occurs during drying, any techniques used to standardize the modified clay pots must
take into account this reduction in size.
The CARE/Kenya team says that during drying the height of the modified clay pot will
come down 2-3 cm, and that the outer circumference of the belly also shrinks. In order to better
understand the extent of shrinkage, a line test was performed at Oriang, and measurements of
freshly made pots after several days of drying were taken at Oriang. Line tests are commonly
used to determine the shrinkage of clay bodies (Kaplan, 2003).
To perform a simple line test, a 9-inch (22.86 cm) long line was drawn on a thin,
rectangular slab of wet clay using a sharp edge. Exactly one day later, the length of the line
reduced to 8-7/8 inches (22.54 cm) long. After the ninth day of drying, the line on the clay slab
measured 8-7/16 inches (21.43 cm) long (-6% linear shrinkage from when molded to when
completely dry). Another line test was performed on a second rectangular slab of wet clay. Only
two measurements were taken: when the line was drawn, it measured 10-2/16 inches (25.72 cm)
long; after the eighth day of drying, the line measured 9-9/16 inches (24.29 cm) long (-6% linear
54
shrinkage from when molded to when completely dry). These results are summarized in Table 61. It is unknown how this linear shrinkage corresponds to volume shrinkage. Norton (1974) notes
that it is often desirable to convert linear shrinkage to volume shrinkage or vice versa.
According to Norton (1974), the ratio is 1 to 3 when changes are quite small; however, as they
get larger, there is substantial departure from this simple ratio.
Table 6-1. Shrinkage line tests performed at Oriang.
Length of line in inches (length of line in cm in parentheses)
Clay slab #1
Clay slab #2
Start = Day 0
9.0000 (22.86)
10.1250 (25.72)
Day 1
8.8750 (22.54)
Day 8
Day 9
8.4375 (21.43)
9.5625 (24.29)
Freshly made pots were also measured after several days of drying to record reduction in
the following pot dimensions: Inner Mouth Diameter, Mouth Circumference, Middle
Circumference (i.e. circumference of the widest part of the belly), Base Circumference, Mouth
Inflection Point, Height, and Height to Middle Circumference. The data for pot measurements at
Oriang on the day that the pots were made as well as seven days later are summarized in Table 62.
55
Table 6-2. Dimensions of modified clay pots made at Oriang on Day 1 (freshly made) and Day 8
(after several days of drying).
Measurements (cm)
Day 1
Pot
Inner
Mouth
I.D.
Diameter
Mouth
Circumf.
Middle
Circumf.
Base
Circumf.
Mouth
Inflection
Height
Point
Height to
Middle
Circumf.
R1
10
36
113
78
5.5
41.5
R2
R3
R4
R5
R6
10
10.5
12
12
10
36
39
44.5
42.5
39.5
108
110.5
114.5
121
117
77
77
88
74.5
86
5.5
5.5
4.5
6.5
7.5
41.5
42
40.5
42.5
45.5
20
19
23.5
22
22
Mouth
Inflection
Point
5
5.5
5
5
5
4
Height
Pot
I.D.
R1
R2
R3
R4
R5
R6
Inner
Mouth
Diameter
9.5
10
9
10.5
10
10
Height to
Middle
Circumf.
20.5
17.5
22
16
16
22
19
8
Mouth
Circumf.
37.5
38.5
35
40
39
38.5
Day
Base
Circumf.
Middle
Circumf.
106.5
113
107.5
106
101
105
65
65
66.5
65.5
68
68
36
39
38.5
38
36
37
Table 6-3, based on the data in Table 6-2, shows the percent shrinkage of the modified
clay pots made at Oriang after several days of drying. The values were calculated by first
subtracting the measurements of Day 8 from the measurements of Day 1, then dividing by the
measurements of Day 1, and finally multiplying by 100. Positive values indicate shrinkage
(white); values equal to 0 indicate no change (light gray); and negative values indicate expansion
(dark gray). While nearly all (75%) of the measurements in Table 6-2 show that the dimensions
of the pots experienced shrinkage during the drying process over several days, some
measurements (14%) increased and others (10%) remained the same. Error in measurement is the
most likely explanation for why some of the measurements appeared to increase or stay the same.
Measurements were taken using a seamstress measuring tape, and it was sometimes difficult to
estimate where to take the measurement for the Middle Circumference, Height to Middle
56
Circumference, Mouth Inflection Point, and Mouth Circumference. Measuring the Height and
the Base Circumference were more straightforward. Overall, while the pots shrunk, they do not
shrink significantly, with an overall average shrinkage of just under 10% (standard deviation is
6%). The 6% linear shrinkage resulting from the two line tests at Oriang seems to agree with this
finding. From what was gathered from interviews, both the potters and CARE/Kenya
underestimate the shrinkage. Both groups mentioned that the height of the modified clay pot
reduces about 2-3 cm after drying; however, according to data in Table 6-2, the height reduces an
average of about 5 cm.
Table 6-3. Percent (%) shrinkage of modified clay pots made at Oriang after several days of drying. Positive
values indicate shrinkage (white); values equal to 0 indicate no change (light gray); negative values indicate
expansion (dark gray).
Pot
1.D.
R1
R2
R3
R4
R5
R6
Avg
Stdev
Inner
Mouth
Mouth
Diameter
5
Circumference
Percent (%) s rinkage in dimensions
Base
Mouth
Middle
Inflection
Circumference
617
Circumference
Point
9
0
16
14
13
17
0
8
7
10
10
8
3
3
7
3
7
17
10
6
7
14
26
9
21
17
6
57
9
r
23
47
13
20
Height to
Middle
Height
Circumference
13
6
8
6
15
19
11
5
13
;
32
27
0
8
19
According to Norton (1974), moving air serves a dual role in the drying process, i.e. "it
supplies heat to the ware as compensation for the evaporative cooling, and it carries away the
water vapor formed. Water evaporated from a piece of ware by drying must come mainly from
the interior of the piece through the fine interconnecting channels." Furthermore, the rate of
evaporation from the surface of drying clay depends on the rate of evaporation from a free water
surface (which is dependent on air temperature, air velocity, and water content of the air) and
drying rates of ceramic bodies, which are explained in detail in Norton (1974):
Should the weight of a drying ceramic piece be plotted against time, the result
would be a smooth curve without any great significance. However, if the rate of
drying, or the slope of the weight-loss curve, is plotted against the water content
of the piece, a curve like that in (Figure 6-6) results. As first pointed out by
Sherwood (1929) [under laboratory conditions], the wet clay starting at point A
dries at a constant rate until B is reached. This constant drying rate is about half
that for a free water surface. At point B the rate starts to decrease rapidly and
finally reaches the origin. It is significant that point B is that at which the mass
changes from a dark to a light color. In other words, down to point B there is over
the surface a continuous water film which acts as free water, but below this point
the water retreats further and further into the pores so that the drying rate becomes
less and less. These steps in the clay structure are shown by (Figure 6-7).
58
B
A
Oi
z
ULU
W-
DC
10
0
20
30
PERCENT MOISTURE
Figure 6-6. Rate of water loss in drying moist clay.
(Source: Norton, 1974)
(a)
.
I
(b)
- .
(d)
(c)
Figure 6-7. Stages in drying moist clay. A cross section at the surface. The letters correspond to those on the
curve of Figure 6-6.
(Source: Norton, 1974)
59
Too much shrinkage may lead to cracking and distortion of a clay piece. The most
common way to reduce drying shrinkage is to add non-plastics such as grog, which is a gritty
material such as ground clay that has already been fired, such that they "reduce the number of
water films per unit distance by displacing a group of clay particles and their associated films by
an equal volume of stable particles" (Norton, 1974). Non-plastics that are relatively coarse
materials will reduce more water films per unit distance; similarly, coarse-grained clays shrink
less than fine-grained clays.
Shrinkage may also occur during the firing process. However, no attempt was made to
calculate percent shrinkage after firing during field work in Kenya since both the potters and
CARE/Kenya stated that the pots only shrink during drying. According to Kingery (1960),
before clay is fired, it is composed of individual grains separated by between 25 and 60 volume
per cent porosity, depending on the particular material and processing method used. During the
firing process this porosity is removed, such that the volume firing shrinkage is equal to the pore
volume eliminated. Adding non-shrinking material, such as grog (pre-fired clay), to the clay
before firing can reduce shrinkage (Kingery, 1960).
60
6.5
Variability in pot dimensions and volumes
Measurements of completed 20 L and 40 L modified clay pots that were for sale at the
collectives were taken using a seamstress measuring tape (Figure 6-8) at all three field site
locations. The following dimensions were measured for each pot: Height, Base Circumference,
Middle Circumference, Height to Middle Circumference, Mouth Circumference, Inner Mouth
Diameter, and Mouth Inflection Point. Table 6-4 displays the dimensions of 20 L modified clay
pots, and Table 6-5 displays the dimensions of 40 L modified clay pots.
Figure 6-8. Measuring pots with a seamstress tape at Amilo.
(Source: Pihulic, 2005b)
61
Table 6-4. Dimensions (cm) of 20 L modified clay pots.
Pot
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Average
Stdev
Variability
Height
38.0
43
42
39
41.5
38
37
41.5
43
44
39
45.5
42
38.5
37
35.5
39
41.5
41
40
36
40
42
43
39
42
41
40.5
44.5
42.5
42.5
41.5
41.5
42
40.5
42.5
45.5
38
37
38.5
47
40.8
2.7
7%
Base
Circumf.
62
61
58.5
60
66
66
74
60
64
60
71
66.5
72
65
65
61
77
69
64
88
56.5
80
85
81
83
81
83.5
87
75
80.5
53.5
78
77
77
88
74.5
86
82
71
54.5
56
71.2
10.4
15%
Middle
Circumf.
114.5
118
111.5
111.5
110.5
106.5
110.5
118.5
116
112
114
118.5
119
111
110
110.5
114
118
113
115.5
109
114
114
123
107
115
114
124
121
115
113
113
108
110.5
114.5
121
117
104
107
112.5
108
113.6
4.6
4%
Height
Middle
Circumf.
21
22
20
18
18
17
18
19
19
20
17
23
20.5
19.5
19
18
20
23
20
19.5
17.5
20
19
19.5
17
22
21.5
18
28
22.5
24
19
20
19
23.5
22
22
17.5
18.5
21
30
20.3
2.7
14%
62
Mouth
Circumf.
37.5
38
34.5
37.5
33.5
35
40
41
34
32
34.5
40.5
38
35.5
33.5
33
32.5
29.5
37
42
34
41.5
43.5
42
42.5
38
41.5
60.5
35
39
36
36
36
39
44.5
42.5
39.5
37.5
37
34.5
33.5
37.9
5.1
13%
Inner
Mouth
Diameter
10
10
9
9
9
9.5
10
11.5
8.5
8
9
11
10
8.5
8.5
8.5
8
7.5
10
10
8.5
11.5
11.5
12.5
11.5
10.5
11
17
9
10
9.5
10
10
10.5
12
12
10
10
10
8.5
7.5
10.0
1.7
17%
Mouth
Inflection Pot
Point
I.D.
6.5
7.5
11
9
10
8
6.5
9
9.5
11.5
8
10
10
9
7
6.5
7
8.5
7
5
2
5
2
7
12
8
7
5
5.5
5
5.5
6
4.5
9.5
5
5.5
6
5.5 R1
5.5 R2
5.5 R3
4.5 R4
6.5 R5
7.5 R6
7.5 F25
4.5
M
6
7.5
7.2
2.1
29% 1
Location
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Amilo
Amilo
Asembo
Asembo
Table 6-5. Dimensions (cm) of 40 L modified clay pots.
Height
Pot
number
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
Average
Stdev
Variability
Height
50
49.5
51
53.5
48
44
46
45.5
56
51.5
55
54
55
55.5
54
53
51
54.5
57
58.5
53.5
50
51.5
51.5
58.5
50
51.5
50
49.5
49.5
51.9
3.6
7%
Inner
Mouth
Base
Middle
Middle
Mouth
Mouth
Inflection
Pot
Circumf.
74
71
76
81
79
75.5
77.5
72
87
74.5
77.5
72.5
82
83
86
78.5
82
79
60
82
85
83
73
66
82
70
79
71
60
48
75.6
8.6
11%
Circumf.
137.5
135.5
152
153
145
127
131
126
138
141.5
148.5
144
154.5
142
139
139
135
141.5
147.5
145
142
139
140
136
148
138
140
138
134
135
140.4
6.9
5%
Circumf.
24
26
27.5
25
22
22.5
25.5
23
33
28
28
33
32
31
29
31
28
28.5
30
32
28.5
29
29.5
34.5
39
30
26
26
29
30
28.7
3.7
13%
Circumf.
34.5
34.5
37
14
32.5
33
33
34
42.5
38
39
38.5
37
43
40.5
37.5
38
36
39
44.5
42.5
37
40.5
43
40.5
37.5
38
36.5
36
35.5
37.1
5.4
15%
Diameter
9
8.5
8
10
8
8.5
9
8.5
12
9.5
10
10
9.5
10.5
10.5
10
10
10
9.5
11
10.5
10
10.5
11.5
11
10.5
10.5
9.5
8.5
Point
8.5
9.5
6
9.5
6.5
6.5
7.5
7
6
7.5
7.5
7.5
6
5.5
5.25
5
5.5
7
7.5
7
6
5
6
6
6.5
5
6
6.5
7.5
7.5
6.7
1.2
18%
1.D.
9.8
1.0
10%
2
2
2
Location
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Oriang
Amilo
Asembo
Asembo
From Table 6-4 and Table 6-5, it is evident that the pot dimensions do not vary much. For
the 20 L modified clay pots, the variability in dimensions is within 17% (average dimensional
variability is 12%), with the exception of the Mouth Inflection Point. The variability in the
Mouth Inflection Point is irrelevant, since the pot is filled with water up to this point; the height
63
above this point does not matter. For the 40 L modified clay pot, discounting the Mouth
Inflection Point dimension since it does not play a role in volume variability, the variability in
dimensions is within 15% (average dimensional variability is 10%). Figures 6-9 and 6-10 are
graphical representations of Tables 6-4 and 6-5, respectively. When comparing pots made within
locations (Table 6-6), average variability in the 20 L and 40 L modified clay pot dimensions are
less than or equal 12% (average dimensional variability is 10%). This suggests that potters
within locations generally shape pots in the same way. However, when comparing pots made
between locations, variability appears much greater. When comparing just the Base
Circumference of 20 L pots between locations, the range is from 55.3 cm (Asembo) to 79.4 cm
(Oriang). This suggests that potters between locations do not generally shape pots in the same
way.
Dimensions of 20 L modified clay pots
140
120AAA
AA AA
A
A
AAAAAAA
AA
* Height
A,
AAAAAAAAAAA
A
AAAAA
A
A AA
a Base Circumference
0 100A
80
No m
Middle Circumference
60 -=moma==
x Height to Middle
Circumference
K Mouth Circumference
40 -
e Inner Mouth Diameter
20
XXX
XX XXXxT
XXX
20 - -xxxx ,xxxxgkxxoxxxxxxxxxx-xxk - "xxxXX
X-X _
+ Mouth Inflection Point
0
0
10
20
30
40
Modified clay pots
Figure 6-9. Graphical representation of the dimensions of 20 L modified clay pots.
64
Dimensions (cm) of 40 L modified claypots
180
160
* Height
A
140
0
A
A
120
A
A
AA
A
* Base Circumference
A AAA
A
A Middle
(n
C
4)
100
4,
80
E
1~
U,
(U
4)
Circumference
x Height to Middle
Circumference
K Mouth Circumference
U
U"
* Inner Mouth Diameter
60
+ Mouth Inflection Point
40
EUX
20
Ac~X
xx x x
X .x
x
XX
~X
Xxxxx
XX
U
x
X X x
xxKX
*+
*
Xx
XX
0
42
52
62
72
Modified clay pots
Figure 6-10. Graphical representation of the dimensions of 40 L modified clay pots.
65
Table 6-6. Variability in 20 L and 40 L modified clay pot dimensions by location.
Height
to
Location
and
Pot size
Amilo
20 L
Amilo
40 L
Oriang
20 L
Oriang
40 L
Asembo
20 L
Asembo
40 L
Average
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Inner
Mouth
Height
39.9
Base
Circ.
66.9
Middle
Circ.
112.7
Middle
Circ.
19.3
Mouth
Circ.
36.0
Mouth
Diam.
9.3
Inflection
Point
8.2
2.7
7%
48.6
3.0
6%
41.9
1.6
4%
53.8
2.6
5%
42.8
6.0
14%
49.5
0.0
0%
7.8
12%
75.2
3.6
5%
79.4
8.0
10%
78.0
7.1
9%
55.3
1.1
2%
54.0
8.5
16%
4.1
4%
138.3
9.9
7%
115.3
4.9
4%
142.0
4.9
3%
110.3
3.2
3%
134.5
0.7
1%
1.7
9%
24.6
1.9
8%
21.1
2.7
13%
30.5
3.0
10%
25.5
6.4
25%
29.5
0.7
2%
3.2
9%
32.1
7.0
22%
41.1
6.0
15%
39.6
2.5
6%
34.0
0.7
2%
35.8
0.4
1%
1.0
11%
8.8
0.7
8%
11.2
1.9
17%
10.4
0.7
6%
8.0
0.7
9%
8.5
2.0
25%
7.5
1.3
18%
5.8
1.2
21%
6.2
0.9
15%
6.8
1.1
16%
7.5
0.0
0%
Overall
Average
11%
10%
12%
8%
10%
3%
The potters at Amilo are each assigned a number so that they can sign the pots that they
make. In this way it is possible to know which pots are made by each potter; however, not all of
the pots are signed. Nonetheless, there is some data available in order to indicate the variability
in pots made by the same potter. Figure 6-11 shows the variability in 20 L modified clay pots
made by Amilo Potter No. 7, while Figure 6-12 shows the variability in 40 L modified clay pots
made by Amilo Potter No. 2. The graphs show that the potters are both very consistent in making
pots.
66
Variability in 20 L modified clay pots' dimensions made by
Amilo Potter No. 7
* Height
120
"
100
a Base Circumference
A
Middle Circumference
x Height to Mddle
Circumference
Mouth Circumference
U) 40
2 Inner Mouth Diameter
0
+ Mouth Inflection Point
0
4
3
2
1
Modified clay pots
Figure 6-11. Variability in 20 L modified clay pots made by Amilo Potter No. 7.
Variability in 40 L modified clay pots' dimensions made by Amilo
Potter No. 2
160 E 140
120100
E
Height
mBase Circumference
A
A
-
80 60 40 20
Middle Cicumference
x Height to Middle Circumference
.
Mouth Circumference
e Inner Mouth Diameter
+ Mouth Inflection Point
x
00
1
3
2
Modified clay pots
Figure 6-12 Variability in 40 L modified clay pots made by Amilo Potter No. 2.
67
To get a preliminary understanding of volume variability, four 20 L modified clay pots at
Amilo were filled with water to the Mouth Inflection Point to determine the exact volume of
water each pot could hold. The volumes were the following: 19 L, 19.3 L, 22 L, and 20 L. The
average of these volumes is 20.1 L with a standard deviation of 1.4 liters, giving a variability of
7%. The dimensions of these four 20 L modified clay pots at Amilo were also measured. These
data are summarized in Table 6-7.
Table 6-7. Measured volumes (L) and dimensions (cm) of four 20 L pots.
Measured
Volume
(Liters)
19
19.3
22
20
20.1
1.4
7%
Pot
number
Al
A2
A3
A4
Average
Stdev
Variability
Height
39
41
41
38
39.8
1.5
4%
Base
Circ.
62.5
58
62.5
70
63.3
5.0
8%
Middle
Circ.
114
110.5
119
115
114.6
3.5
3%
Height
to
Middle
Circ.
21
23
21
17
20.5
2.5
12%
Mouth
Circ.
37.5
34.5
38
34.5
36.1
1.9
5%
Inner
Mouth
Diameter
10
9.5
10
9
9.6
0.5
5%
Mouth
Inflection
Point
8
10.5
7.5
7.5
8.4
1.4
17%
Location
Amilo
Amilo
Amilo
Amilo
In an attempt to relate dimensional variability to volume variability, a method was
developed to convert known pot dimensions to volumes. The following symbols (Figure 6-13)
were used, with units in brackets:
VTOT =
Vb =
Total Volume [cm 3
Volume of the bottom [cm 3
Vt = Volume of the top [cm 3]
r = radius [cm]
z = height [cm]
rb =
radius of the bottom (i.e. Base Circumference / 27r) [cm]
68
rm = maximum radius (i.e. Middle Circumference / 27E) [cm]
rt = radius of the top (i.e. Mouth Circumference / 2n) [cm]
h = height from rb to rm (i.e. Height to Middle Circumference) [cm]
h' = height from rm to rt (i.e. Height - h - Mouth Inflection Point) [cm]
a = value relating r to z
1
Vt
Vb
Figure 6-13. Symbols used in equations to convert dimensions to volumes.
The following equations were used to approximate the radius, r, and the Total Volume, VTOT
1)
VTOT= Vb + Vt
2) Vb
=
Jar 2 dz
(integrating from 0 to h)
3) r = rb + rm - rb(z/h)' = rb + ((rm
4) Vb
= 7 [rb2 h + (2rb((rm - rb)Ih
-
rb)Ih a)z "
(substitute r into Equation 2)
c)h a+')/(a+1) + (((rm
-
rb)Ih a) 2 h 2 a+1)/(2a+1)]
5) Vt = n [rt2h' + (2rt((rm - rt)/h' )h ' "*)/(a+1) + (((rm - rt)/h' a) 2 h' 2 a+1)/(2a+1)]
69
Equation 1 gives VTOT in cubic centimeters; therefore, to covert VTOT to liters, it is necessary to
divide by 1000 (since 1000 cm 3 = 1 L). The value that relates r to z, a, was calibrated using the
known volumes and dimensions of the four 20 L modified clay pots found in Table 6-7. The a
values ranged from 0.64 to 0.89, with an average a = 0.79. This average value of a was used to
back-calculate the volumes of the four 20 L modified clay pots found in Table 6-7. Table 6-8
compares the measured volumes with the calculated volumes. While the average of the
calculated volumes is equal to the average of the measured volumes, the variability in the
calculated volumes (9%) is slightly greater than the variability in the measured volumes (7%).
Table 6-8. Comparison of measured volumes to calculated volumes of four Amilo 20 L modified clay pots.
Measured
Volume
(Liters)
Calculated
Volume
(Liters)
Al
19
19.7
A2
19.3
18.1
A3
22
22.6
20
20.1
1.4
7%
19.8
20.1
1.9
9%
Pot
number
A4
Average
Stdev
Variability
The key assumptions made in developing this method (Equations 1-5) for converting pot
dimensions to volumes are that the shapes of the modified clay pots are all uniform and that the a
for Vt is the same as the a for Vb. However, as can be seen from Figure 6-14, the shapes of
modified clay pots are not uniform, and the way that the radius relates to the height (i.e. the
curvature of the pot, or a) differs for the tops and bottoms of the pots. It was hypothesized that a
would differ between locations and between differently sized pots (20 L vs. 40 L). Thus, this
method, which should be taken as an approximate method to convert pot dimensions to pot
volumes, was applied to calculate volumes from the dimensions of additional 20 L modified clay
70
pots from only the Amilo location (Pot numbers 1-21, 38, and 39 found in Tables 6-4), since a 20
L Amilo modified clay pot was used to calibrate a. These calculated volumes are found in Table
6-9. For the 20 L Amilo modified clay pots, the average volume is 20.0 L with a variability of
13%. These data agree with the data in Table 6-8. When this method was used to calculate the
average volumes from the dimensions of all other modified clay pots found in Tables 6-4 and 6-5
(Pot numbers 22-37 and 40-7 1), the results (Table 6-10) varied widely, suggesting that a is
indeed both location-specific and size-specific. Thus, in order to apply this method to a wider
range of modified clay pots, further research is required so that the a value can be calibrated
separately with both measured dimensions and measured volumes for each size and for each
location.
The volumes of modified clay pots are important since chlorine dosing is prescribed for
20 L volume multiples. The CDC determined that a roughly 10% volume variability is
acceptable to dose the volume of water with the prescribed amount of chlorine (CDC, 2004).
However, since people who own modified clay pots generally use 20 L plastic jerry cans,
buckets or some other standardized container to transport water from sources to fill the modified
clay pots, generally do not fill the modified clay pots completely (i.e. to the Mouth Inflection
Point), and generally dose the water with chlorine in the containers used to transport the water
before filling the modified clay pots, the issue of volume variability does not play such an
essential role in chlorine dosing as previously thought (Lantagne, 2005). However, little
variability in volume is still a valid goal, and this thesis showed that the volume variability in 20
L Amilo modified clay pots is under 10%, when averaging the measured volume variability of
Pot numbers A1-A4 (Table 6-8), the calculated volume variability of Pot numbers A1-A4 (Table
6-8), and the calculated volume variability of Pot numbers 1-21, 38, and 39 (Table 6-9).
71
Figure 6-14. Varying shapes of modified clay pots.
(Source: Murcott, 2004; Pihulic, 2005b; Murcott, 2004)
Table 6-9. Calculated volumes (VTOT in L) of Amilo 20 L modified clay pots.
Pot
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
38
39
Average
Stdev
Variability
rt
6
6
5.5
6
5.3
5.6
6.4
6.5
5.4
5.1
5.5
6.4
6
5.7
5.3
5.3
5.2
4.7
5.9
6.7
5.4
6
5.9
rm
18.2
18.8
17.7
17.7
17.6
17
17.6
18.9
18.5
17.8
18.1
18.9
18.9
17.7
17.5
17.6
18.1
18.8
18
18.4
17.3
16.6
17
rb
9.9
9.7
9.3
9.5
10.5
10.5
11.8
9.5
10.2
9.5
11.3
10.6
11.5
10.3
10.3
9.7
12.3
11
10.2
14
9
13.1
11.3
h
h'
a
21
22
20
18
18
17
18
19
19
20
17
23
20.5
19.5
19
18
20
23
20
19.5
17.5
17.5
18.5
10.5
13.5
11
12
13.5
13
12.5
13.5
14.5
12.5
14
12.5
11.5
10
11.5
9
14
13.5
14
8.5
10.5
13
14
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
0.79
72
Location
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
Amilo
VTOT
20.1
23.4
18.3
17.9
18.8
17
19.5
21.5
21.3
19.3
19.9
24.6
22.8
18.2
18.2
16
22.7
24.7
21.1
21.6
15.7
18.5
19.2
20.0
2.5
13%
Table 6-10. Calculated volumes (VTOT in L) of modified clay pots by location and size.
Location and
Calculated volumes (L)
using method
size of pot
Amilo
20 L
Average
Stdev
Variability
Amilo
Average
40 L
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Average
Stdev
Variability
Oriang
20 L
Oriang
40 L
Asembo
20 L
20.0
2.5
12.5%
36.9
6.4
17.2%
25.4
2.6
10.2%
46.1
4.8
10.4%
20.7
2.3
11.3%
Asembo
Average
33.1
40 L
Stdev
Variability
1.2
3.5%
73
6.6
Techniques and tools to standardize pottery
There are several techniques and tools available to standardize handbuilding, which is
constructing pottery from pre-made parts that are molded, coiled, or fashioned by hand. As
mentioned in section 6.2, the modified clay pot is constructed from coils that are placed one on
top of the other. The most common tools used to standardize pottery making are in general
molds, calipers, and measuring tapes. With the exception of one potter from the Oriang
Women's Pottery Group, none of the potters used any of these tools.
A mold is a plaster shape designed to pour slip cast into and let dry so the shape comes
out as an exact replica of the mold (Stoke-on-Trent, 2005). Slip refers to clay mixed with water
to make it fluid-like, and cast means to produce shapes by pouring this fluid clay into molds.
Molds are usually made from plaster of Paris, otherwise known as gypsum or calcium sulphate
(Pottery Studio, 2001). In more general terms, a mold is a hollow form or matrix for shaping a
2
fluid or a plastic, or a frame or model around or on which something is formed or shaped . It
may be possible to produce a mold for the modified clay pot; however, due to the large size of
the pot (20 L or 40 L) and the limited resources for materials, this technique is not the most ideal
method for standardization.
Calipers are used to make pots similar in size because they measure inside and outside
dimensions, or, more specifically, the diameter of round forms. For example, calipers are used to
get lids to fit just right (Stoke-on-Trent, 2005). Figure 6-15 shows examples of both hardwood
and aluminum calipers.
2 http://dictionary.com
74
Figure 6-15. Hardwood calipers (left) and aluminum calipers (right) used to standardize pottery.
(Source: BigCeramicStore.com, 2005)
Measuring tapes can also be used to control size, although they are a rather simple and
unsophisticated tool. However, their ease of use is ideal for low-technology areas such as rural
districts of developing countries. When making modified clay pots, potters would have to use the
measuring tape frequently in order to control the size and ensure proper dimensions. The tool
actually tested during field work in Kenya is an even simpler version of a measuring tape: a
piece of rope with 3-4 knots marking off certain dimensions of the modified clay pot. These
ropes are termed "reference ropes".
6.7
Reference Ropes
Reference ropes were developed to help standardize the dimensions of the modified clay
pot. This simple tool consists of a length of rope with knots tied to indicate measurements that
correspond to dimensions of the modified clay pot. It was thought that a minimum of three
dimensions (i.e. Height to Middle Circumference, Middle Circumference, and Inner Mouth
Diameter) would be necessary to make the desired modified clay pot, and that the other correct
75
dimensions would naturally be achieved by the potter's experience and skill. A reference rope
for a 20 L modified clay pot was made based on the following calculations, taking into account
10% shrinkage in volume during the drying process:
Volume of 20 L modified clay pot approximated as sphere =
V = 4/3 nTr 3 = 20 L = 20000 cm 3
Radius = r = 16.8 cm
Circumference = C = 2nr = 105.8 cm
10% of volume = V 1 0 = 0.1 (4/3 71 (16.8 cm) 3 )
Volume taking into account 10% shrinkage
Vnew=
=
1986.17 cm 3
= Ve
=
20000 cm 3 + 1986.17 cm 3
21986.17 cm 3 = 4/3 nr3
Rnew
= 17.38 cm~ 18 cm =Height to Middle Circumference
Cnew
= 109.20 cm ~ 110 cm = Middle Circumference
It was assumed an Inner Mouth Diameter equal to 10 cm would be both narrow enough to
discourage hand and cup dipping and wide enough to allow attaching of the tap and cleaning of
the pot. Table 6-11 summarizes the three dimensions calculated for the 20 L modified clay pot
reference rope. Figure 6-16 shows a pictorial drawing of this reference rope and Figure 6-17
shows a photograph of the actual reference rope.
Table 6-11. Dimensions calculated for the 20 L modified clay pot reference rope.
Measurement
(cm)
Height to Middle Circumference
Middle Circumference
Inner Mouth
Diameter
18
110
10
76
Inner Mouth Diameter
Height to Middle Circumference = 18 cm
=
10 cm
Middle Circumference = 110 cm
Figure 6-16. Pictorial drawing of 20 L modified clay pot reference rope.
Figure 6-17. Photograph of 20 L modified clay pot reference rope.
A potter at Amilo was the first one to try using the 20 L modified clay pot reference rope
(Figure 6-18). Halfway through making the pot, it was obvious that the pot was too small and it
was impossible to make the Middle Circumference wide enough to match the reference rope's
specifications. It was thought that the potter started with too small of a base. This perhaps is
reason to also mark off a desired Base Circumference on the reference rope in addition to the
other three dimensions. She destroyed the unfinished pot and started over. The second time
around she was encouraged to use the reference rope more frequently to check that the
dimensions would be accurate. With coaching from the researchers (Figure 6-19) and reminders
to keep using the reference rope, the potter succeeded the second time in making a beautiful 20 L
modified clay pot in exact accordance with the dimensions designated on the reference rope. The
potters reported that the ropes were easy to understand and easy to use, two factors essential to
developing a successful standardization tool in this context. The exact shrinkage of the modified
77
clay pot during the drying process will need to be determined to make more accurate reference
ropes, since 10% shrinkage was assumed based on measurements in Section 6.4 and the results
in Table 6-3. A method of trial-and-error problem solving in conjunction with good record
keeping is necessary to improve the reference rope tool and to make a similar reference rope for
the 40 L modified clay pot.
Figure 6-18. A potter using the 20 L modified clay pot reference rope at Amilo.
(Source: Pihulic, 2005b)
Figure 6-19. Teaching the potters how to use the reference rope tool.
(Source: Pihulic, 2005b)
78
6.8
Summary of results
The potters learn to make 20 L and 40 L pots from knowledge passed down and from
experience. They frequently use their own arm lengths or hand widths to know how wide or long
to make certain pot dimensions. Line tests performed at Oriang showed that linear shrinkage of
the clay is about 6% during the drying process. Modified clay pot dimensions measured over
several days showed that freshly made pots at Oriang shrunk an average of 10% during the
drying period.
20 L and 40 L modified clay pots were measured at all three field sites (Amilo, Oriang,
and Asembo) to determine variability in pot dimensions. The average dimensional variability of
the 20 L pots is 12%, while the average dimensional variability of the 40 L pots is 10%. When
comparing pots made within each location, the average dimensional variability is 10%,
suggesting that potters within locations generally shape pots in the same way. However, when
comparing pots made between each location, the variability appears much greater, suggesting
that potters between locations do not generally shape pots in the same way. Data also showed
that potters are individually very consistent at making similarly sized clay pots. After measuring
the volumes of 20 L modified clay pots at Amilo and calculating their volumes using a method to
convert pot dimensions into volumes, it was found that the volume variability is within 10%.
This is acceptable to dose the volume of water with the prescribed amount of chlorine (CDC,
2004). Further research is required to convert pot dimensions into volumes for a wider range of
modified clay pots by location and size.
Molds, calipers, and measuring tapes are common techniques and tools used to
standardize pottery. For this thesis, a reference rope tool was developed to help control the
dimensions of the 20 L modified clay pot. Specifically, the rope was marked off to specify
79
Height to Middle Circumference, Middle Circumference, and Inner Mouth Diameter. The use of
tools such as reference ropes can be encouraged as a tool for training new potters to make the
modified clay pots; however, given the little variability in pot dimensions and volumes, reference
ropes are not necessary for potters already experienced in making modified clay pots. If the
reference ropes are used, it is important to account for shrinkage that will occur during the
approximately two-week drying period.
80
Chapter 7
Tap design and attachment
The current tap design, the metal tap, represents a significant investment, costing about
the same as all other materials required to manufacture the modified clay pot combined. When
also factoring in the materials required to attach the tap, i.e. sand, regular cement, waterproof
cement, and the black PVC pipe, the cost is even higher, accounting for approximately 60-65%
of the total cost of the finished pot. Also, because the preparation and attachment of the metal tap
to the modified clay pot wall is one of the most complex stages of production, requiring the
assembly of several individual components, the time to attach the tap is lengthy. In addition, the
techniques for attaching the tap also have not been perfected and often create leaks that reduce
the effectiveness of the pots' ability to store water. Finally, the method of attaching the metal tap
to the pot using cement is such that it is difficult to replace the tap if it breaks.
Because of these cost concerns and leakage problems, one goal of this thesis has been to
design a replacement tap that will accomplish, as a minimum, two things: 1) reduce the cost of
the modified clay pot; and, 2) eliminate or reduce the leakage from the tap attachment area. This
chapter describes the current design and attachment process, characterizes the leakiness of this
current design, and discusses alternative designs. Then it describes the recommended new design,
the plastic tap, and finally summarizes the findings.
81
7.1
Current design and attachment
The current design used to dispense water from the modified clay pot is the metal tap. It
consists of a threaded shaft on one end, an outlet through which water flows downward on the
other end, and a handle that must be turned on a horizontal plane to control the flow of water.
The taps used by the three different pottery groups cost between 100-200 KSH (US$1.30US$2.70 3). Both
and 3/4 in. diameter taps are used, where the diameter refers to the diameter of
the outlet of the tap through which the water flows downward. Locals prefer the
in. diameter
taps because the flow rate of water is increased. The outer diameter of the threaded shaft is
typically
in., onto which a PVC pipe with 7/8 in. inner diameter and 1 in. outer diameter is
attached to extend its length before attaching it to the pot. Figure 7-1 shows an example of a
metal tap.
HANDLE
BODY
IN. DIAMETER
OPENING
2 IN.
DIAMETER
OPENING
IN. LONG THREADED
SHAFT
1/2
Figure 7-1. Example of a metal tap attached to modified clay pots and used to safely dispense water.
3 Exchange rate is 75 KSH = US$1.00
82
Before the metal tap is attached to the modified clay pot, it must be prepared. This tap
preparation process is described in detail in Pihulic, 2005a:
A tap must first be modified lightly by extending the length of its body before it
can be attached to the vessel. Currently all three groups use a variable length of
PVC tubing with 7/8 in. inner diameter and 1 in. outer diameter. Extension of the
tap length is required to accommodate the physical space needed to interact with
and operate the tap and to secure a sufficiently strong join between the tap and the
vessel. The tap is attached near the base of the vessel to maximize the amount of
the water that is accessible. Curvature near the base of the vessel combined with
the vessel's thickness, approximately 1 cm to 1.5 cm, require the tap's threaded
end to be longer in order to install the tap level to the ground. Additionally, taps
with levers for handles cannot function if they are installed too close to the vessel
surface. Strength requirements are also an issue. Because the tap extends beyond
the surface of the vessel it is prone to be jostled, bumped or even used as a grip. A
minimal amount of extension beyond the vessel interior and exterior [is] required
in order to properly cement the fixture in place. The size of the PVC [pipe]
extension is variable, usually ranging from 6 cm to 10 cm.. .The pipe is then cut
using a hacksaw blade (Figure 7-2).. .Once the tubing has been cut to the proper
length it must be fitted to the tap joint.. .The Oriang and Amilo groups use a crude
form of press fitting to attach the tubing to the tap, [i.e.] a small fire is used to
heat the end of the pipe (Figure 7-3). The pipe is then pressed over the threading
of the tap joint. This process is repeated until the pipe fits snugly over the threads
(Figure 7-4). On the final pass the Oriang potters coat the threads in a rubber
cement compound to ensure a good seal is formed. In contrast, the Kinda E Teko
group cuts a cm vertical notch in the end of the pipe and press fits it to the tap
joint (Figure 7-5). The notch relieves tension allowing the pipe to spread to
accommodate the threading... Once the tube is attached, the tap is ready to be
joined to the vessel.
Figure 7-2. Oriang potter cutting PVC pipe to proper length with a hacksaw blade.
(Source: Pihulic, 2005a)
83
Figure 7-3. Oriang potter heating PVC pipe to thermally press fit it to the metal tap threads.
(Source: Pihulic, 2005a)
Figure 7-4. The PVC pipe fits snugly over the metal tap threads after thermal press fitting.
(Source: Pihulic, 2005a)
Figure 7-5. Close-up of split PVC pipe used by Kinda E Teko potters to fit tap to PVC pipe.
(Source: Pihulic, 2005a)
84
The method of using cement to attach the prepared metal tap to the modified clay pot is
similar among the three pottery groups. Because the Oriang group attaches the tap in the most
time-efficient manner, their technique is described below (as found in Pihulic, 2005a):
A single potter works on several vessels in parallel (Figure 7-6). To prevent
cement from clogging the tap during fixing, the end of the PVC tube is stuffed
with plastic or cardboard. The vessels are placed on a bench, and the [prepared]
tap is inserted into the opening. The taps are then wedged into the openings, using
[fired pottery] shards and vessels to prop up the tap. Because the opening is made
near the base where the surface is tilted, wedging the tap in the opening only
produces a minor tilt, allowing it to remain roughly parallel with the ground
(Figure 7-7). Several vessels are secured in succession. A 1-to-1 mixture of
cement and sand referred to as "nill", a local term for the substance, is mixed with
some water and used to fix the tap in place. Attachment of the tap occurs in two
stages over the course of two days. In stage 1 only the opening and a small region
around it in the exterior and interior are covered in nill. This initial deposit of nill
serves to stabilize the tap and fix it in place (Figure 7-8). Once it has been allowed
to dry for a day, and has been checked for cracks, a covering layer of nill is
applied to the base (Figure 7-9) and the lower surface of the vessel, extending up
to the tap (Figure 7-10). The vessel is placed upside down. The additional
covering layer serves two purposes. First it serves to strengthen the base and the
tap attachment. Second it serves to cover up cracks that may have formed in the
base or initial layer of nill. Overall the potter may spend two hours working on
three pots over the course of two days.
Figure 7-6. Tap attachment in parallel at Oriang.
(Source: Pihulic, 2005a)
85
Figure 7-7. Tap wedged in vessel using another clay vessel and shards for stabilization and support.
(Source: Pihulic, 2005a)
Figure 7-8. Tap is fixed in place using nill on inside of outside of vessel.
(Source: Pihulic 2005a)
Figure 7-9. Nill is applied to the base, forming an outer layer.
(Source: Pihulic, 2005a)
86
Figure 7-10. Nill is extended up to the tap and around the base of the vessel.
(Source: Pihulic, 2005a)
One difference to note with regards to the attachment of the tap at the Amilo CBO pottery
group is that rocks and pottery shards are used to stabilize the tap in the opening and wedge it
into place (Figure 7-11). Cement is then used to fill in the cracks. It is thought that this method
contributes to the likelihood that the area around the tap will leak.
Figure 7-11. Using rocks and pottery shards to stabilize tap and wedge it into vessel at Amilo.
(Source: Pihulic, 2005a)
87
7.2
Leakiness from current tap
Leaking in the modified clay pot is caused by high porosity levels in the clay material,
the formation of cracks in the pot wall, and poor attachment of the metal tap to the pot (Pihulic,
2005a). The porosity of the clay allows for the flow of water through the pot wall and its
evaporation from the exterior surface, leading to an evaporative cooling effect (Schiffer, 1990).
This evaporative cooling effect is one reason locals like the modified clay pots. Cracks in the pot
wall caused by the stresses of drying, firing, or poor shaping can also contribute to leaking.
Frequent opening and closing of the metal tap by turning its handle can generate stresses that
many loosen the tap if it is not properly secured during the attachment process. These stresses
may create cracks and cause leaking (Pihulic, 2005a). Based on interviews with the potters, an
excessively leaky pot is defined by a more than one third (33%) volume loss of water over 24
hours. Similarly, an excessively leaky tap is one that contributes to a water volume loss of
greater than one third (33%) over 24 hours.
In order to test leakiness from the modified clay pots, both from the pot as a whole and
from the area around the metal taps, experiments were conducted at Oriang and Amilo. At
Oriang, three 40 L modified clay pots were filled with exactly 40 L of water using a 1 L plastic
measuring cup. The volume of water was then measured the following day, about 24 hours after
the pots were first filled, by emptying out water from the tap one liter at a time. Because one of
the pots had a substantial crack around the tap, water leaked out at the rate of about 150 ml per
minute (9 liters per hour). At this rate the pot would be empty in less than 5 hours. When it
became clear that the leaking would continue at this rate, the pot was emptied. This is an
example of an excessively leaky tap. Table 7-1 shows the results of the leakiness experiment for
the other two pots filled at Oriang. One pot lost 7% of its volume, while the other pot lost 17% of
88
its volume. This loss of water was due to a combination of the pot's porosity and leakage around
the tap. Thus, these two Oriang pots are not excessively leaky.
A second leakiness experiment was conducted at Amilo by filling four 20 L modified
clay pots up to their respective Mouth Inflection Point. The results of this experiment are
summarized in Table 7-2. The volume of water lost over 24 hours ranges from 6% to 32%. The
average water loss for all the pots filled, including those from both Oriang and Amilo, is 15%.
Thus, besides the one Oriang pot that leaked excessively from the tap area, none of the pots
studied leaked excessively. Despite this finding, the potential for leaking from the metal tap area
is high because of the way the metal tap is attached using cement and because of the stresses
caused by frequent turning of the tap handle to dispense water.
Table 7-1. Leakiness experiment at Oriang.
Pot
number
01
02
Volume water (Liters)
Day 2
Day 1
37.3
40
33.2
40
Water loss
7%
17%
Table 7-2. Leakiness experiment at Amilo.
Pot
number
Al
A2
A3
A4
Volume water (Liters)
Day 2
Day 1
17.9
19
15.2
19.3
15.8
22
18.5
20
89
Water loss
6%
21%
32%
9%
7.3
Alternative designs
Several ideas were generated as an alternative to the current metal tap design. They are
described below:
(a) Attach calabash (ladle) to the inside of the pot by hooking it under the lid (Figure 7-12a);
(b) Plug the hole with a rubber stopper, removing it to dispense water (Figure 7-12b);
(c) Place rubber washers on metal tap currently used instead of using cement (Figure 7-12c);
(d) Construct wooden tap similar to those found on European wooden beer casks (Figure 7-12d);
(e) Insert plastic tap with rubber washers, using jam nut to secure it to pot wall (Figure 7-12e);
(f) Insert a tube in the hole and regulate flow with a clamp (Figure 7-12f).
d
c
b
a
f
e
Figure 7-12. Alternative designs to current metal tap design: attached ladle (a); rubber stopper (b);
rubber washers on metal tap (c); wooden tap (d); plastic tap (e); tube with clamp (f).
(Source: American Science & Surplus, 2005 (b); African Trading Company, 2004 (d); Tomlinson
Industries, 2001 (e); McMaster-Carr, 2005 (f))
90
A Pugh chart was used to compare each of these alternative designs to the current design.
This tool helps to evaluate ideas by setting up a list of characteristics and judging each idea in
terms of the design criteria. The current metal tap design was chosen as the datum, or the idea to
which all others were compared. For each criterion, it was decided if the option being evaluated
was the same (0), better (1), or worse (-1) than the datum. The results for each are totaled to
determine which idea is the best. The criteria used to evaluate alternative tap designs were:
-
Cost;
-
Leakiness;
-
Access to materials;
-
Ease of use;
-
Ease of construction or attachment;
-
Social acceptability;
-
Contamination likelihood; and
-
Maintenance or fixability.
The Pugh chart is shown in Table 7-3. The plastic tap scored the highest, with a total of 4, while
the rubber stopper scored the second highest, with a total of 2. The plastic tap was ultimately
chosen to be tested in the field as an alternative to the metal tap.
91
Table 7-3. Pugh chart used to evaluate tap design alternatives.
Evaluation criteria
Ease
TOTAL
Cost
Leakiness
Material
access
of
use
Ease of
construct
Social
accept
Contamination
Maint.
or fix
Metal tap (datum)
Attached calabash
0
1
0
1
0
0
0
-1
0
0
0
0
0
-1
0
1
0
1
Rubberstopper
Rubber washers +
metal tap
Wooden tap
Plastic tap
Tube with clamp
1
1
-1
0
1
0
-1
1
2
-1
1
1
1
1
0
1
0
-1
0
-1
-1
0
0
1
-1
0
-1
1
0
0
0
0
-1
0
0
0
-1
1
1
1
1
0
1
4
-2
Options
7.4
Plastic tap
The plastic taps tested in the field were purchased from a water cooler store in the United
States. The manufacturer of these plastic taps (HFSA series, Model #1000019) is Tomlinson@
Industries, which is located in Cleveland, Ohio. The plastic taps are similar to the ones used in
the Potters for Peace Filtr6n System (Figure 7-13). Ron Rivera of Potters for Peace has
experimented with an assortment of spigots in the field and has found that the spring-operated
plastic spigot "shows [more] promise for overall acceptance and practical usability" than
alternate designs (PFP 2001). Field experience in Nicaragua showed Rivera that target
households consistently preferred manufactured taps over taps "hand fashioned from readily
available materials such as bamboo, flexible hoses, and PVC tubing" because of their perceived
inferior quality, no matter their actual durability (PFP 2001).
The Tomlinson@ plastic taps tested in the field consist of a threaded shaft onto which two
rubber washers and a jam nut are secured. The water is dispensed by pushing down a springloaded lever that opens the tap. The threaded shaft is 1 in. long, has a 9/16 in. inner diameter and
an 11/16 in. outer diameter. However, the diameter including the threads is 3/4 in. The outlet
92
through which the water flows has a 3/8 in. inner diameter. The rubber washers have % in. inner
diameters and 1-1/4 in. outer diameters. They are flat on one side and slanted on the other side.
Figure 7-14 shows a schematic drawing of the plastic tap.
The cost of each Tomlinson@ Industries plastic tap is US$4.68 (350 KSH) when
purchased from the manufacturer. If the plastic faucet, washers and jam nut are each bought
separately in bulk, then the prices of each are US$1.35 (101 KSH), US$0.19 (14 KSH) and
US$0.26 (20 KSH), respectively. Thus, the total cost of the plastic tap (including two washers) is
US$1.99 (150 KSH). Incidentally, Lantagne (2005) has a special deal with Tomlinson@
Industries in which the plastic taps only cost US$0.75 (56 KSH) each. She purchases these taps
for the Jolivert Safe Water for Families Project in Haiti.
Figure 7-13. Potters for Peace Filtr6n System, with a spring-operated plastic tap.
(Source: PFP, 2001)
93
WASHERS
HANDLE
diameter
opening
3/4 in.
diameter
THREADS
JAM NUT
BODY
PARTITION
(e.g. pottery
wall)
3/8 in.
diameeter
openi ng
1 in. long THREADED SHAFT
Figure 7-14. Schematic drawing of the spring-operated plastic tap.
No preparation is needed before attaching the plastic tap to the modified clay pot wall.
However, the hole must be small enough so that the rubber washers fit snugly and create a watertight seal with the pot wall. In addition, the pot wall cannot be too thick; otherwise, the jam nut
cannot be screwed on and secured tightly. The pot wall is typically 1-1.5 cm (0.4-0.5 in.) thick,
while the shaft length is 1 in. long. The following steps are required to attach the plastic tap: first,
the jam nut and one of the washers is removed; then, the plastic tap is placed into the hole from
the outside of the pot; and finally, the rubber washer and jam nut are replaced from the inside of
the pot. There is one rubber washer on either side of the pot wall (Figure 7-14).
The first attempt to attach the plastic tap to a pot occurred at Oriang. All of the finished
modified clay pots could not accommodate the plastic tap because the holes were too large.
Potters typically make the holes larger than the diameter of the PVC pipe attached to the metal
tap since the current attachment process involves using cement to fill in the extra space between
94
the pipe and pot wall. Several pots with holes the right sizes were still drying and would be fired
in a few days. Therefore, other options were considered for the first trial test. A small flower pot
was found to have a hole nearly the right size for the tap to fit through, i.e. slightly bigger than
the threaded diameter of the plastic tap. Since the hole was still a bit too large, an additional large
rubber washer (1/8 in. thick, 2-1/4 in. outer diameter, inner diameter cut to fit the threaded
diameter) was placed on the inside of the pot. Unfortunately, the rubber washer was unable to fit
snugly to the pot wall because there was a slight lip (raised edge) around the hole on the inside of
the pot. When filled with water, the area around the tap leaked a lot. When the large rubber
washer was moved to the outside of the pot (Figure 7-15), which had a flat surface around the
hole and no lip, there was no leakage because the seal was good. Figure 7-16 shows a close-up of
the jam nut used to secure the tap on the inside of the small flower pot at Oriang.
Figure 7-15. The plastic tap attached to a small flower pot at Oriang. An additional large rubber washer was
placed on the outside of the pot to create a better seal, since the hole was slightly too large for the plastic tap.
(Source: Pihulic, 2005b)
95
Figure 7-16. Close-up of the jam nut used to secure the plastic tap on the inside of the small flower pot at
Oriang.
(Source: Pihulic, 2005b)
The second attempt to attach the plastic tap occurred at Amilo. A few 20 L modified clay
pots with holes that were only slightly bigger than the plastic tap's threaded diameter were
chosen to test the plastic tap. The additional large rubber washer was again used to since the
holes were still too large. Overall, the roundedness of the pots did not allow for good seals.
However, on the third pot tried, there was no leakage.
The third and final attempt to attach the plastic tap occurred at Oriang and involved using
the recently fired 20 L modified clay pots with holes that were made with the plastic tap itself, to
ensure the correct size. It is important to note that each of these modified clay pots had a unique
improvement, i.e. a flat area around the holes to create a better fit with the tap. Though shrinkage
was considered, the holes shrank more than hypothesized. The edge of a pair of heavy duty
scissors was used to widen the holes by scraping away some of the fired clay around the hole.
The plastic tap was then attached to one of these 20 L modified clay pots, filled with water, and
observed for leakage (Figure 7-17). There was no leakage evident.
96
Figure 7-17. A 20 L modified clay pot at Oriang with the plastic tap attached. No leakage was observed from
the tap area.
(Source: Pihulic, 2005b)
One advantage of the plastic tap over the metal tap is that the spring-operated tap
automatically shuts the tap after each use, thereby making it more user friendly, especially for
children. In addition, the spring-loaded handle on the plastic tap creates less stress on the pot
than the metal tap, which has a metal lever that must be turned 90' counter-clockwise on a
horizontal plane by using a good amount of force and sometimes requiring a free hand to steady
the pot. Since the rubber washers and jam nut are sold together with the plastic tap, no additional
materials, e.g. cement or PVC pipe, are required for attachment. Another huge advantage is that
the plastic taps can be removed and replaced safely without destroying the pots. Since the
threaded shaft of the metal tap is only
in. long, compared to the 1 in. long threaded shaft of the
plastic tap, washers and a metal jam nut cannot feasibly be added to the metal tap. Finally, the
cost of the plastic tap is US$0.75 (56 KSH), which is less than the cost for a metal tap (100-200
KSH). This price, however, is a lower bound estimate.
97
It is unknown whether the spring-operated plastic tap is available in local Kenyan
markets. An initial search located one on a water purifying unit sold in the capital city of Nairobi;
however, none were found to be sold separately. CARE/Kenya suggested that perhaps
Kentainers, a local plastic manufacturer well-known for making large plastic water storage
containers, could be contacted to determine their interest and the feasibility in producing plastic
taps. Lantagne (2005) has visited all the plastic manufacturers in Kenya and discovered none are
currently capable of making the plastic taps. However, Lantagne (2005) learned similar springoperated plastic taps (Indian-manufactured) are imported from India to Nigeria, where they retail
for approximately US$1.00 (75 KSH). Since Kenya has as good, if not better, contacts with India
than Nigeria, and all the plastic manufacturers in Kenya are Indian owned and operated (and it
costs less to transport taps from India to Kenya than from India to Nigeria), perhaps the best
estimate on a price for spring-operated plastic taps in Kenya is equal to or less than US$1.00 (75
KSH). On a related note, Murcott (2005) learned that spring-operated plastic taps are being
imported from China to Ghana by a ceramic water filter manufacturer, on the advice of Ron
Rivera of Potters for Peace. These plastic taps, which are Chinese-manufactured and imported to
Ghana, sell for US$2.25 (169 KSH) each in Ghana. This compares with the metal taps in Ghana,
which were US$4.00 (300 KSH) in the remote market in the north (3/4 in. size), and were
US$2.00 (150 KSH) in the capital, bought in bulk (
in. size). It may be possible, then, for
Kenya to import plastic taps from India or China and that the costs would be similar to those in
Nigeria and Ghana, respectively.
98
The following improvements could be made in the design of the spring-operated plastic
tap:
(1) a longer threaded shaft to better account for the thickness and roundedness of the
pot;
(2) a larger jam nut to distribute force over a greater area of the pot wall and be easier
to grasp; and,
(3) the diameter of the outlet through which water flows can be made larger so that
flow rates are increased.
One improvement that can be made in the manufacture of the modified clay pot is to mold a flat
area around the tap hole to create a better fit with the tap.
7.5
Summary of results
The current metal tap design requires substantial preparation before attachment, and the
tap attachment process itself is lengthy and tedious. There is high interest to replace the metal tap
with a cheaper design that can be removed and replaced if necessary.
Leakiness experiments conducted at Oriang and Amilo showed that pots on average lost
about 15% of their volume of water over 24 hours. This loss of water is due to a combination of
factors, including water escaping through the pores (the evaporative cooling effect of the clay),
leakage from cracks in the pot, and leakage from the tap area. According to potters, excessively
leaking pots are characterized by a greater than one third volume loss over 24 hours. The
potential for leaking from the tap area is high because of the way the metal tap is attached using
cement and because of the stresses caused by frequent turning of the tap handle to dispense water.
99
A Pugh chart was used to evaluate several alternative designs to the metal tap. Based on
cost, leakiness, material access, ease of use, ease of construction, social acceptability,
contamination, and maintenance or fixability, the plastic tap was found to be the most promising
design to replace the metal tap. However, the plastic tap itself may need some modifications,
including a longer threaded shaft, a larger jam nut, and a larger outlet for which water to flow
through.
The plastic tap is spring-operated and similar to the ones found on the Potters for Peace
Filtr6n System. The attachment process is simple and does not require any preparation. A plastic
tap attached to a 20 L modified clay pot at Oriang did not leak, showing must promise. The
plastic tap has several advantages over the metal tap, including its lower cost, elimination of
leakage, ease of use, ease of attachment, and its capacity to be removed and replaced without
ruining the entire pot. Thus, it satisfies the two minimum requirements outlined in the beginning
of this chapter. Further research is required to determine its local availability in Kenyan markets
and the possibility of importing it from elsewhere.
100
Chapter 8
Cost Recovery Analysis
Because one of the goals of the program in Kenya is to reach those most in need, and the
more affordable the pot the more people can be reached, cost recovery of all the materials, labor
and transportation used to manufacture the modified clay pot is a key component. The Bovin and
Morohashi (2002) Best Practicesstudy reports that the metal tap used to dispense water in the
modified clay pot is very expensive, costing nearly the same amount as all of the other materials
combined. There are some variations in the material inputs necessary to manufacture the
modified clay pots at each field site. For example, potters at Kinda E Teko do not mix sand into
the clay because the clay is of a better quality than that found at Amilo and Oriang; therefore,
sand does not factor into the cost breakdown of the modified clay pots made at Kinda E Teko in
Asembo. Below is a list of all of the material inputs and their purposes:
-
Clay: Molded to form pots;
-
Sand: Mixed in with the clay before molding pots, and also mixed with regular
and waterproof cements and used to cover bottom of pot and attach tap;
-
Pala: Decorative clay used on the outside of pots;
-
Red oxide: Decorative red coloring used on the outside of pots;
-
Dried grass: Used to cover pots and to start the fire during the firing process;
-
Small wood: Used to cover pots and to fuel the fire during the firing process;
-
Metal tap: Used to safely dispense water from modified clay pots;
-
Black PVC pipe: Attached to the metal tap to make it longer and thus more easily
fitted into the hole cut out specifically for the metal tap in the modified clay pot;
101
-
Regular cement: Mixed with waterproof cement and sand and used to cover
bottom of pots and to attach tap;
-
Waterproof cement: Mixed with regular cement and sand and used to cover
bottom of pots and to attach tap.
It is important to distinguish material costs from other costs, including labor costs (payment to
the potters), transportation costs (costs of transporting pots to and from the market), and sale
prices of the pots. In this thesis, material cost is taken to be the cost of each material item,
including any labor and transportation costs required to obtain the item. It is assumed that the
breakdown of costs at all three locations does not apply specifically to either 20 liter- or 40 litersized modified clay pots, but to an overall average of all the modified clay pots produced at that
location.
8.1
Cost breakdown of Amilo modified clay pot
Data for the breakdown of material costs used to manufacture the modified clay pots
produced at Amilo CBO in Rangwe were obtained by interviewing people at the Amilo CBO site
(Figure 8-1) and are summarized in Table 8-1 and Figure 8-2. The situation quickly became
somewhat chaotic because everyone wanted to contribute to calculating costs of each material
input, i.e. ten people would discuss the bulk price of each component and give a rough estimate
of the unit cost based on the number of pots each bulk quantity of materials could yield.
According to their numbers, grass and small wood make up nearly three-quarters of the total
material cost of the modified clay pot, which seems a large overestimation, considering these
material inputs are used just to fire the pots. The total cost of one modified clay pot, based on
this estimate, equals 1018 KSH (US$13.60), which is more than twice the sale price of a 40 L
102
modified clay pot, which is 500 KSH (US$6.70). The sale price of a 20 L modified clay pot is
only 380 KSH (US$5.10). Costs for transporting the pots to and from the market are unknown.
According to Amilo CBO members, payment to potters is 150 KSH (US$2.00) for making a 40
L modified clay pot and 100 KSH (US$1.30) for making a 20 L modified clay pot. These data
are questionable since the numbers clearly do not add up.
Figure 8-1. Interviewing Amilo CBO members for cost breakdown of a modified clay pot at Amilo.
(Source: Pihulic, 2005b)
103
Table 8-1. Material cost breakdown of Amilo modified clay pot (reported by Amilo CBO).
Cost
per item
Number of pots
Cost
per pot
(KSH)
made per item
(KSH)
Clay
Sand
Pala
2000
3500
2000
50
50
500
40
70
4
Red oxide
250
20
13
Grass
25800
50
516
86 bundles of grass
Small wood
Metal tap
Black PVC pipe
Waterproof cement
10500
150
200
250
50
1
61
20
210
150
3
13
Includes labor and transportation
Larger tap costs 180 KSH
Cost for 20 feet; 10 cm used per pot
Item
Notes
Cost is for area of land that contains
clay
Includes driver fee and vehicle rental
1018
Total Material Cost
Material cost breakdown of Amilo Modified Clay Pot
(reported by Amilo C.B.O.)
Waterproof
cement '
Black PVC pipe
1%
0%
Clay
4%
Sand
7%
o Clay
o Sand
* Pala
Metal tap
15%
Small wood_,
21%
Pala
0%
Red oxide
1%
o Red oxide
* Grass
o Small wood
Grass
51%
" Metal tap
" Black PVC pipe
o Waterproof cement
Figure 8-2. Pie chart showing material cost breakdown of Amilo modified clay pot, as reported by
Amilo CBO members.
104
In order to double check these cost data, CARE/Kenya staff member Alex Mwaki was
contacted to provide a more accurate breakdown of costs (Mwaki, 2005). These data,
summarized in Table 8-2 and Figure 8-3, seem more reasonable, as they show that the metal tap
comprises about half the total material cost of the modified clay pot, which is consistent with the
researcher's understanding from the Bovin and Morohashi (2002) report. According to Mwaki's
(2005) numbers, the total material cost of the Amilo modified clay pot is now approximately 370
KSH (US$4.90), which is almost identical with the sale price for the 20 L modified clay pot,
which is 380 KSH (US$5.10). However, when adding in the labor cost for a 20 L modified clay
pot, which is 100 KSH (US$1.30), the cost of the Amilo modified clay pot is 470 KSH
(US$6.30). Thus, based just on Mwaki's (2005) material costs plus labor costs for a 20 L
modified clay pot, cost recovery is not achieved. Similarly, when adding the labor cost for a 40 L
modified clay pot, 150 KSH (US$2.00), to the material costs, the cost of the Amilo modified clay
pot is 520 KSH (US$6.90). The sale price of the 40 L modified clay pot is 500 KSH (US$6.70).
Thus, cost recovery is again not achieved. Table 8-3 summarizes the material costs, labor costs,
transportation costs and sale prices for 20 L and 40 L modified clay pots at Amilo.
Transportation costs are not included because they are unknown. It is also unknown whether or
not CARE/Kenya subsidizes any of the material, labor or transportation costs. Furthermore, it is
unknown who coordinates payment to the potters, and it may be possible that the women potters
are subsidizing some of the costs of the modified clay pots through unpaid labor. Further
research is required to identify unknown transportation costs and subsidies to determine whether
cost recovery is indeed not achieved.
105
Table 8-2. Material cost breakdown of Amilo modified clay pot (reported by Mwaki, 2005).
Cost
Cost
per item
(KSH)
20
800
Item
Clay
Sand
Number of pots
made per item
1
181
per pot
(KSH)
20
4
Pala
10
1
10
Red oxide
Grass
Small wood
120
90
20
15
1
1
8
90
20
Metal tap
Black PVC pipe
Regular cement
Waterproof cement
200
120
630
120
1
60
50
50
200
2
13
2
Total Material Cost
Notes
Estimate
Sold in tons, 5 kg/pot, 1 ton = 907 kg
3/4 in. size
369
Material cost breakdown of Amilo Modified Clay Pot
(reported by Mwaki, 2005)
Waterproof
cement
Regular cement
3%
1%
Cl
5
% Sand
0
1%
o Clay
Pala
3%
Black PVC pipe
Re d oxide
1%
2%
Grass
24%
Metal tap
55%
Small wood
5%
0 Sand
U Pala
0 Red oxide
* Grass
a Small wood
" Metal tap
* Black PVC pipe
" Regular cement
" Waterproof cement
Figure 8-3. Pie chart showing material cost breakdown of Amilo modified clay pot, as reported by
Mwaki, 2005.
106
Table 8-3. Summary of material costs, labor costs, transportation costs and sale prices for 20 L and 40 L
modified clay pots at Amilo.
Amilo modified clay pots
8.2
20 L
40 L
Material cost
Labor cost
Transportation cost
370
100
?
370
150
?
Total cost
470
520
Sale price
380
500
Cost breakdown of Oriang modified clay pot
Data for the breakdown of material costs used to manufacture modified clay pots
produced at the Oriang Women's Pottery Group in Oriang were obtained by interviewing
members of the Oriang Women's Pottery Group and CARE/Kenya staff and are summarized in
Table 8-4. However, this material cost breakdown is incomplete since Oriang members and
CARE/Kenya staff could not recall the prices of all of the material items. When the potters at
Oriang first started making modified clay pots in 2000 (see section 1.5 in Chapter 1),
CARE/Kenya provided metal taps, black PVC pipe, waterproof cement and regular cement
without charge since these items were the additional ones necessary to modify the clay pots to
have taps. While the Oriang Women's Pottery Group now say it is financially independent, it is
unclear who is responsible for buying certain components used to make the modified clay pots.
However, if one uses the itemized costs of materials provided by Mwaki (2005) for Amilo (Table
8-2) for those items for which material costs were not available for Oriang (a fair assumption
given the groups operate in the same general locale and purchase materials locally), the total
material costs of the unknown materials would equal roughly 60 KSH (US$0.80). This brings the
total material cost of the Oriang modified clay pot to about 270 KSH (US$3.60). This
assumption gives a material cost that falls in the range of the sale price for the 20 L modified
107
clay pot, which is between 250-280 KSH (US$3.30-$3.70), and for the 40 L modified clay pot,
which is 350 KSH (US$4.70). However, the labor costs and transportation costs are unknown, so
it is unclear whether cost recovery is actually achieved. Furthermore, the sale prices for 20 L and
40 L modified clay pots at Oriang differ enormously from those at Amilo. The Oriang potters
reported that they originally sold the 40 L modified clay pot for 500 KSH (US$6.70) but that
CARE/Kenya told them to reduce the price because it was too high. Further research is required
to determine whether or not CARE/Kenya subsidizes some of the material, labor, and
transportation costs at Oriang, causing the Oriang potters to not be commercially independent but
still very much viable as a social enterprise. Figure 8-4 shows a graphical breakdown of material
costs of the Oriang modified clay pot, assuming that the unknown material costs equal 60 KSH
(US$0.80). Table 8-5 summarizes what is known of the material costs, labor costs, transportation
costs and sale prices of 20 L and 40 L modified clay pots at Oriang.
Table 8-4. Material cost breakdown of Oriang modified clay pot (reported by Oriang potters and
CARE/Kenya staff).
Item
Cost
per item
Number of pots
Cost
per pot
(KSH)
made per item
(KSH)
200
400
150
10
10
1
20
40
150
Notes
Clay
Sand
Pala
Red oxide
Grass
Small wood
Metal tap
Black PVC pipe
Regular cement
Waterproof cement
210+
Total Material Cost
108
Taps range from 150-200 KSH
Material cost breakdown of Oriang modified clay pot
(reported by Oriang members and CARE/Kenya)
Grass
7%
Other
22%
Small wood
15%
m Grass
o Small wood
m Metal spigot
o Other
Metal spigot
56%
Figure 8-4. Pie chart showing material cost breakdown of an Oriang modified clay pot, as reported
by the Oriang potters and CARE/Kenya staff. Note that an assumption for the unknown costs of some
materials ("other") was made based on data from Table 8-2.
Table 8-5. Summary of material costs, labor costs, transportation costs and sale prices for 20 L and 40 L
modified clay pots at Oriang.
Material cost
Labor cost
Transportation cost
Total cost
Sale price
Oriang modified clay pots
40 L
20 L
270
270
?
?
?
?
270
270
250-280
350
109
8.3
Cost breakdown of Kinda E Teko (Asembo) modified clay pot
Data for the cost breakdown of materials used to manufacture modified clay pots
produced at Kinda E Teko Pottery Group in Asembo were obtained by Gibney (2005) by
interviewing people at the Kinda E Teko site and are summarized in Table 8-6 and Figure 8-5.
The total cost of the Kinda E Teko modified clay pot is 202 KSH (US$2.70), and the sale price is
370 KSH (US$4.90). It appears that cost recovery is achieved; however, labor and transportation
costs are unknown. Table 8-7 summarizes what is known of material costs, labor costs,
transportation costs and sale prices of modified clay pots at Kinda E Teko.
Table 8-6. Material cost breakdown of Kinda E Teko modified clay pot (reported by Kinda E Teko
potters to Gibney, 2005).
Cost
Cost
per item
(KSH)
Number of pots
made per item
per pot
(KSH)
Clay transport
Red oxide
100
80
4
80
25
1
Small wood
Metal tap
Black PVC pipe
Regular cement
Waterproof cement
450
100
300
600
120
10
1
20
50
30
45
100
15
12
4
Item
202
Total Material Cost
110
Notes
Clay is free, cost is for donkey
transport
Negligible amount used per pot
Material cost breakdown of Kinda E Teko Modified Clay
Pot (reported by Kinda E Teko to Gibney, 2005)
Regular cement
6%
Black PVC pipe
Waterproof
cement
Clay transport
12%
2%
7%
o Clay transport
o3 Red oxide
o Small wood
Red oxide
* Metal tap
0%
* Black PVC pipe
* Regular cement
E Waterproof cement
Small wood
22%
Metal tap
51%
Figure 8-5. Pie chart showing material cost breakdown of a Kinda E Teko modified clay pot, as
reported by the Kinda E Teko potters to Gibney (2005).
Table 8-7. Summary of material costs, labor costs, transportation costs and sale prices for modified clay pots
at Kinda E Teko.
Kinda E Teko modified
clay pots
Material cost
Labor cost
Transportation cost
Total cost
Sale price
202
?
?
202
370
111
8.4
Limitations of analysis
There are several limitations to this cost recovery analysis, mainly due to a lack of
complete financial information:
(1) In order for the cost recovery analysis to be complete, data on material costs,
labor costs and transportation costs is necessary. Labor costs are only available
for Amilo, and transportation costs are unknown for all three pottery sites.
(2) It is unknown whether or not CARE/Kenya subsidizes any of the material, labor
and transportation costs at Amilo and Oriang and whether or not SWAK
subsidizes any of the material, labor and transportation costs at Kinda E Teko. It
is also unknown whether the women potters subsidize any of these costs through
unpaid labor.
(3) It is unclear how each of the pottery groups sets the sale prices of its modified
clay pots, i.e. whether or not sale prices are set at or near cost (including materials,
labor and transportation) of the modified pots and whether or not there are markups in sale prices to include a profit margin, and if so, who is collecting that profit.
(4) It is unknown how loss of pots due to breakage during transportation, cracking or
other means factors into the cost recovery analysis.
(5) Due to the lack of complete data necessary to determine cost recovery, the
financial viability and hence sustainability of the pottery groups' enterprises
cannot be clearly ascertained.
112
8.5
Summary of results
It is unknown whether cost recovery is achieved at all three pottery locations, due to
incomplete or unknown data on material, labor and transportation costs, subsidies, setting of sale
prices and profit margins, and losses due to pot breakage, pot leakage, or other means. The
understanding that the metal tap comprises about half of the total material cost of the modified
clay pot is confirmed at all three locations (Table 8-8).
Table 8-8. Summary of the cost of the metal tap, total material cost of modified clay pot and the
metal tap's percentage of the total material costs at Amilo, Oriang, and Kinda E Teko.
Cost of metal tap
Total material costs of modified clay pot
Metal tap cost % of total material costs
113
Amilo
200
370
54%
(All prices in KSH)
Kinda E Teko
Oriang
100
150
202
270
50%
56%
Chapter 9
Conclusions
9.1
Standardization
Contrary to perceptions that the modified clay pot sizes and shapes are non-standard, this
thesis shows that variability in both 20 L and 40 L pot dimensions and volumes is within an
acceptable range, and that potters are both individually and collectively consistent in making
similarly sized and shaped clay pots. One concern of the rounded shape of the pots is that it is
difficult to transport. While no attempt was made to address this issue, findings from focus
groups conducted at Amilo and Oriang showed that potters seem to want to retain the rounded
shape of the pots. While it is possible that a cylindrical-shaped pot is more easily transported,
further research is required to determine its feasibility and acceptability.
For this thesis, a reference rope tool was developed to help control the dimensions of the
20 L modified clay pot. The use of tools such as reference ropes can be encouraged as a tool for
training new potters to make the modified clay pots. However, given the little variability in pot
dimensions and volumes, reference ropes are not necessary for potters already experienced in
making modified clay pots. The data suggests that the current method of passing down
knowledge and learning by experience is sufficient. If the reference ropes are used, it is
important to account for shrinkage that will occur during the approximately two-week drying
period. More research is necessary to better quantify this shrinkage.
114
9.2
Tap design and attachment
The current metal tap design requires substantial preparation before attachment, and the
tap attachment process itself is lengthy and tedious. The current design is also prone to leakage,
is expensive, and cannot be removed and replaced if it becomes defective. The spring-operated
plastic tap was found to be the most promising design to replace the metal tap for several reasons:
1) no preparation required before attachment;
2) simple attachment process;
3) lower cost;
4) the spring-operated tap automatically shuts the tap after each use, thereby making it
more user friendly, especially for children;
5) no leakage;
6) can be removed and replaced without ruining the entire pot; and,
7) the spring-operated handle creates less stress on the pot.
Further research is required to determine if it works in households and whether or not it is locally
available in Kenyan markets. If the plastic taps cannot be found locally, CARE/Kenya suggested
that perhaps Kentainers, a local plastic manufacturer, could be contacted to determine their
interest and the feasibility in producing plastic taps. Perhaps a more promising and immediate
option is to determine how to import spring-operated plastic taps from India or China, similar to
what is being done in Nigeria and Ghana, respectively.
Several improvements could be made in the design of the spring-operated plastic tap.
First, a longer threaded shaft could better account for the thickness and roundedness of the pot.
Second, a larger jam nut would distribute force over a greater area of the pot wall and be easier
to grasp. Third, the diameter of the outlet through which water flows downward can be made
115
larger so that flow rates are increased. One improvement that can be made in the manufacture of
the modified clay pot is to mold a flat spot around the tap hole to create a better fit with the tap.
9.3
Cost recovery
Due to incomplete or unknown data on material, labor and transportation costs, subsidies,
setting of sale prices and profit margins, and losses due to pot breakage, pot leakage, or other
means, it is unknown whether cost recovery is achieved at all three pottery locations. Sufficient
data was available to determine a complete breakdown of material costs at both Amilo and Kinda
E Teko, but not at Oriang. The understanding that the metal tap comprises about half of the total
material costs of the modified clay pot is confirmed at all three locations.
9.4
Overall conclusions
This thesis documented, analyzed, and suggested improvements for the design of the
modified clay pots, specifically with regards to the standardization of the size and shape of the
pots and the tap design and attachment. Further research is required to determine whether or not
cost recovery of the modified clay pot is achieved at all three pottery sites visited in Kenya. It is
hoped that this information will be useful to potters, NGOs, and others committed to making
water safe through disinfection and safe storage at the point of use.
116
References
African Trading Company (2004). Close-up of the wooden spigot.
http://www.sciplus.com/category.cfm?category=50 Accessed 25-6-2005.
American Science & Surplus (2005). Lab supplies and equipment-corks, stoppers and lids.
http://www.sciplus.com/category.cfm?category=50 Accessed 25-6-2005.
BigCeramicStore.com (2005). Kemper calipers.
http://www.bigceramicstore.com/Supplies/Kempertools/kemper calipers.htm Accessed
6-6-2005.
Bovin K (Nuffic) and Morohashi J (UNESCO) (2002). Editors. Best practicesusing indigenous
knowledge - Africa. "The use of locally produced clay pots modified for safe storage of
drinking water in the home-a component of CARE Kenya's Nyanza Health Water
Project." A joint publication of Nuffic, The Hague, The Netherlands, and
UNESCO/MOST, Paris, France.
CARE Kenya (2003). A manual on the Safe Water System based on CARE Kenya's experience
in rural western Kenya. FirstEdition. Nairobi, Kenya.
CDC (2004). Chlorine dosing report: Maputo, Mozambique. Centers for Disease Control and
Prevention, Atlanta, GA. 13 pp. Available from:
http://home.earthlink.net/-danielelantagne/Mozambigue%20Stuff/Deliverables/Dosing%
20Report.pdf Accessed 26-4-2005.
CDC (2005). Safe Water Systems for the developing world: A handbookfor implementing
household-basedwater treatment and safe storageprojects.
http://www.cdc.gov/safewater/ Accessed 25-5-2005.
Central Bureau of Statistics (2003). Kenya's CentralBureau of Statistics: Ministry ofplanning
and national development. http://www.cbs.go.ke/ Accessed 28-6-2005.
Dictionary.com (2005). Entriesfor "mold". http://dictionary.reference.com/search?q=mold
Accessed 6-6-2005.
Dies R (2003). Development of a ceramic waterfilterfor Nepal. Available from:
http://ceemeng.mit.edu/Water/public html/Docs/Student%20Theses/Nepal/Dies2003.pdf
Accessed 21-10-2004. Master of Engineering in Civil and Environmental Engineering
Thesis, Massachusetts Institute of Technology, Cambridge, MA.
Eleveld A (2004). "Safe Water Systems as a home care intervention for people living with
HIV/AIDS." Presented at the 2004 International Symposium "Household Technologies
for Safe Water," hosted by the Kenya Ministry of Health, Nursing Council of Kenya,
117
Emory University, Population Services International (PSI), the Centers for Disease
Control and Prevention (CDC) and CARE/Kenya in Nairobi, Kenya on June 16-17, 2004.
FAO (2005). Map of Kenya showing administrativeboundaries. Available from:
http://www.fao.org/ag/agl/swlwpnr/reports/y sf/z ke/ke map/kemp I 12.pdf Accessed
22-6-05.
Gibney J (2005). Notes from field work in Kenya, January2005. Graduate student,
Massachusetts Institute of Technology, Cambridge, MA. Email: igibney@mit.edu
Gordon B, Mackay R, and Rehfuess E (2004). Inheritingthe world: The atlas of children's
health and the environment. World Health Organization (WHO). Available from:
http://www.who.int/ceh/publications/en/atlas.pdf Accessed 25-5-2004.
Hwang R (2003). 2003 Six-month field monitoring of point-of-use ceramic waterfilter by using
H2S paper strip most probable number method in San FranciscoLibre, Nicaragua.
Available from:
http://ceemeng.mit.edu/Water/public html/Docs/Student%20Theses/Nicaragua/Hwang20
03.pdf Accessed 21-10-2004. Master of Engineering in Civil and Environmental
Engineering Thesis, Massachusetts Institute of Technology, Cambridge, MA.
Kaplan J (2003). PPP:Testing for success. Available from:
http://www.ceramicindustry.com/CDA/Articlelnformation/features/BNP
m/0,2710,104732,00.html Accessed 24-6-2005.
Kenya Web (2001). Nyanza Province map.
http://www.kenyaweb.com/regions/nyanza/nyanza.html
Features
Ite
Accessed 25-5-2005.
Krueger R (1994). Focus groups:A practicalguide for applied research. Second Edition. Sage
Publications, Inc. Thousand Oaks, California.
Kingery W (1960). Introduction to ceramics. John Wiley & Sons, Inc. New York.
Lantagne D (2001a). Investigation of the Pottersfor Peace colloidal silver impregnatedceramic
filter. Report 1: Intrinsic effectiveness. Alethia Environmental. Allston, Massachusetts.
Lantagne D (2001b). Investigation of the Pottersfor Peace colloidal silver impregnatedceramic
filter. Report 2: Field investigations. Alethia Environmental. Allston, Massachusetts.
Lantagne D (2005). Personalcommunication with Young, S. on June 27, 2005 via email.
Daniele Lantagne, PE, Centers for Disease Control and Prevention, Foodborne and
Diarrheal Diseases Branch. Email: dlantagne@cdc.gov
Low C (2002). Appropriate microbialindicatortestsfor drinking water in developing countries
and assessment of ceramic waterfilters. Available from:
http://ceemeng.mit.edu/Water/public html/Docs/Student%20Theses/Nepal/Low2002.pdf
118
Accessed 21-10-2004. Master of Engineering in Civil and Environmental Engineering
Thesis, Massachusetts Institute of Technology, Cambridge, MA.
McMaster-Carr (2005). Plasticflow-control tube clamps. http://www.mcmaster.com/ Accessed
25-6-2005.
Murcott S (2004). Photographsfromtrip to Kenya, summer 2004. Susan Murcott, Lecturer,
Massachusetts Institute of Technology, Cambridge, MA. Email: murcott@mit.edu
Murcott S (2005). Personal communication to Young, S. on June 24, 2005 via email. Susan
Murcott, Lecturer, Massachusetts Institute of Technology, Cambridge, MA. Email:
murcott@mit.edu
Mutagaywa J (1995). "Pottery technology in Kenya." Do it herself: Women and technical
innovation. Ed. Appleton H. Intermediate Technology Publications, London, UK. 162166.
Mwaki A (2005). Personalcommunication with Young, S. on May 5, 2005 via email. Alex
Mwaki, Staff, CARE/Kenya Homa Bay Office. Email: alex@ksm.care.or.ke
Norton F (1974). Elements of ceramics: Second Edition. Addison-Wesley Publishing Company.
Reading, MA.
Pihulic M (2005a). An analysis of the production and manufacture of the modified clay pot at
the Oriang Women's Pottery Group, the Amilo-Rangwe Pottery Group, and the Kinda E
Teko Pottery Group in Nyanza Province, Kenya. Master of Engineering in Civil and
Environmental Engineering Thesis, Massachusetts Institute of Technology, Cambridge,
MA.
Pihulic M (2005b). Photographsfrom Master of Engineeringthesisfield work in Kenya,
January2005. Graduate student, Massachusetts Institute of Technology, Cambridge,
MA. Email: pihulic@alum.mit.edu
PFP (2001). Filters:technical data: spigots. Potters for Peace.
http://www.potpaz.org/pfptechdata.htm#spigots Accessed 14-6-2005.
Pottery Studio (2001). The Pottery Studio glossary of terms and abbreviations.
http://www.studiopottery.com/news/glossary.html Accessed 6-6-2005.
Quick R, Kimura A, Thevos A, Tembo M, Shamputa I, Hutwagner L, and Mintz E (2002).
Diarrheaprevention through household-level water disinfection and safe storage in
Zambia. American Journal of Tropical Medicine and Hygiene. 66(5):584-589.
Available from: http://www.cdc.gov/safewater/publications pages/2002/guick 2002.pdf
Accessed 25-5-2005.
119
Quick R (2004). Personalcommunication with Young, S. on December 13, 2005 via telephone.
Robert Quick, MD, MPH, Foodborne and Diarrheal Diseases Branch, Centers for Disease
Control and Prevention, Atlanta, Georgia. Email: rxql @cdc.gov
Rice P (1987). Pottery analysis:A sourcebook. Chicago Press; Chicago, IL.
Schiffer M (1990). "Technological change in water-storage and cooking pots: Some predictions
from experiment." The changing roles of ceramics in society: 26,000 B.P. to the present.
Ed. Kingery W. Westerville, OH: American Ceramic Society, Inc. 119-136.
Semenza J, Roberts L, Henderson A, Bogan J, and Rubin C (1998). Water distributionsystem
and diarrhealdisease transmission:a case study in Uzbekistan. American Journal of Tropical
Medicine and Hygiene. 59:941-6. Available from:
http://www.cdc.gov/safewater/publications pages/1998/semenza 1998.pdf Accessed 25-52005.
Sherwood T (1929). The drying of solids. Ind. Eng. Chem. 21(12).
Stoke-on-Trent (2005). A-Z of ceramics. http://www.thepotteries.org/types/az.htm Accessed 66-2005.
Tomlinson Industries (2001). No-drop commercialfaucets: Forhot water, hot and cold
beverages and bottled water coolers.
http://www.tomlinsonind.com/products/nodrip/index.asp Accessed 25-6-2005.
United Nations Development Programme-Kenya/GEF-SGP (2004). Provincialmap of Kenya.
http://www.ke.undp.org/GEF-SGP/Kenya%20project%20frame.htm Accessed 25-52005.
Wandibba S (2003). "Ceramic ethnoarchaeology: Some examples from Kenya." East African
archaeology: Foragers,potters, smiths, and traders. Eds. Kusimba C and Kusimba S.
University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia.
59-70.
Yamana T (2004). Post-filtration:Safe water storage and chlorination. Undergraduate
Research Opportunities Program (UROP) Trip to Dominican Republic Report.
Massachusetts Institute of Technology, Cambridge, MA. Email: tkcy@alum.mit.edu
120